1522
`
`Current Perspectives
`
`Direct Thrombin Inhibitors
`in Cardiovascular Medicine
`
`Jeffrey Lefkovits, MBBS; Eric J. Topol, MD
`
`Currently used antithrombotics such as heparin
`Abstract
`have a number of potential limitations that may be overcome
`by the new class of agents that directly inhibit thrombin. These
`agents variously block the active catalytic and/or the anion
`binding exosites of the thrombin molecule and are potent and
`specific inhibitors of thrombin's many biological actions, as
`demonstrated by in vitro and animal models of thrombosis.
`Preliminary data indicate that the direct antithrombins are
`safe and efficacious in humans, and their use in acute coronary
`
`syndromes and coronary angioplasty in place of heparin has
`yielded promising early results. Phase III trials in these clinical
`settings are currently under way. Newer antithrombotics that
`inhibit thrombin generation and thrombin activity at various
`strategic points within the coagulation cascade are also in the
`early stages of development. (Circulation. 1994;90:1522-1536.)
`Key Words * antithrombotics * thrombosis *
`coagulation
`current perspectives
`
`*
`
`T he introduction of the direct thrombin inhibi-
`tors - a new class of agents that specifically and
`potently antagonize the actions of thrombin -
`has opened up new and exciting opportunities in the
`regulation of the thrombotic process, which is so impor-
`tant in the field of cardiovascular medicine. It is now
`firmly established that coronary artery thrombosis is
`causal in the development of acute coronary syn-
`dromes.12 Thrombin in turn plays a pivotal role in both
`the platelet activation and the fibrin generation inher-
`ent in this process and offers an ideal target for modu-
`lation of the clinical course of these syndromes. Hepa-
`rin, currently the mainstay of antithrombotic therapy,
`has a number of limitations that may be overcome by
`specific characteristics of the newer thrombin inhibitors.
`This review will focus on these agents, from their
`biochemistry through to the current status of their
`clinical use in the cardiological fields of acute coronary
`syndromes and coronary angioplasty. Their use has
`been evaluated in several other areas, including dissem-
`inated intravascular coagulation,3 hemodialysis,4 micro-
`surgery,5 and deep venous thrombosis,6 but will not be
`covered in this review.
`Limitations of Current Antithrombotics
`The principal antithrombotic action of heparin de-
`pends completely on the presence of cofactors. Inacti-
`vation of thrombin; activated factor X (factor Xa);
`factors XII, XI, and IX; and tissue factor Vlla complex7
`is achieved through the activation and modulation of
`antithrombin 111.8 Heparin may also act by potentiating
`heparin cofactor II-mediated inactivation of thrombin,
`especially at higher doses.9 The therapeutic use of
`
`Received February 23, 1994; revision accepted May 18, 1994.
`From the Department of Cardiology, Center for Thrombosis
`and Vascular Biology, The Cleveland Clinic Foundation, Cleve-
`land, Ohio.
`Correspondence to Eric J. Topol, MD, Department of Cardiol-
`ogy, One Clinic Center, Cleveland Clinic Foundation, Cleveland,
`OH 44195.
`© 1994 American Heart Association, Inc.
`
`heparin mandates intensive and meticulous laboratory
`monitoring of its anticoagulant effect because of the
`marked individual variation in its anticoagulant re-
`sponse. Heparin can be inactivated by platelet factor 4
`and heparinase, both of which are released by activated
`platelets,10 and fibrin monomers have been found to
`protect thrombin from inactivation by the heparin-
`antithrombin III complex in thrombogenic states."
`Binding to vitronectin, fibronectin, and other plasma
`proteins limits the amount of heparin available and can
`decrease its anticoagulant effect,12 whereas heparin
`therapy is ineffective in people with hereditary anti-
`thrombin III deficiency. Most important, however, is the
`inability of the heparin-antithrombin III complex to
`inactivate thrombin already bound to clot.1"12 This is
`most likely due to a conformational change in the
`thrombin molecule induced by its binding to fibrin,
`which, in turn, prevents attachment of the heparin-
`antithrombin III complex.13 The "protected status" of
`clot-bound thrombin is a major potential drawback to
`heparin use, as this thrombin can act as an ongoing
`source of thrombogenesis at sites of pathological throm-
`bus formation, theoretically limiting heparin efficacy.
`In contrast, the actions of direct thrombin inhibitors
`are antithrombin III independent; they do not bind
`plasma proteins to a significant extent, are able to
`inactivate clot-bound thrombin, and prevent thrombin-
`induced platelet activation, thus offering several poten-
`tial advantages over heparin. Their mechanisms of
`action and likely therapeutic benefit in cardiovascular
`disorders may be better understood by reviewing the
`biology of thrombin itself.
`Biology of Thrombin
`Thrombin, the key regulator of the thrombotic pro-
`cess, is a glycosylated, trypsinlike serine protease, with
`lysine- or arginine-directed specificity.14 It is generated
`from prothrombin by the prothrombinase complex-
`factors Xa, Va, calcium, and phospholipids, with factor
`Xa cleaving the prothrombin molecule to form the A
`
`
`
`http://circ.ahajournals.org/Downloaded from
`
` by guest on October 7, 2015
`
`CFAD v. Anacor, IPR2015-01776
`ANACOR EX. 2007 - 1/16
`
`

`
`Lefkovits and Topol
`
`Direct Thrombin Inhibitors
`
`1523
`
`Simplified overview of the coagulation cas-
`FIG 1.
`cade, illustrating the central roles of factor Xa and
`thrombin in the coagulation process. The sites of
`action of the various antithrombotic agents are also
`shown. ATIII indicates antithrombin 111; LMWH, low
`molecular weight heparins; PC, protein C; APC, (re-
`combinant) activated protein C; TFPI, tissue factor
`pathway inhibitor; TRAP, thrombin receptor antagonist
`peptide; TAP, tick anticoagulant peptide; and TM,
`thrombomodulin.
`
`--
`
`'bxlla
`
`FR
`
`Protein S /
`
`Acved Protein C
`
`Inactvion Va, Villa
`
`and B chains of thrombin, which combine to form
`a-thrombin.
`Thrombin has many and varied biologic functions, but
`its main action is to catalyze the transformation of
`fibrinogen to fibrin, whether the thrombin is soluble in
`plasma or fibrin bound. Thrombin further activates
`factor XIII to cross-link fibrin and stabilize the clot, and
`promotes and amplifies clot formation by activating
`other clotting plasma proteins, including factors V and
`VIII. Of importance, it also is one of the most potent
`agonists for platelet recruitment and aggregation, which
`further reinforces the newly formed clot15 (Fig 1).
`Thrombin also initiates a series of counterregulatory
`reactions to trigger endogenous anticoagulant systems
`to maintain homeostasis. Through its binding to throm-
`bomodulin, thrombin activates protein C, which joins
`with protein S to inactivate factors Va and VIIIa (Fig 1).
`It also stimulates the release of both tissue-type plas-
`minogen activator (TPA) and plasminogen activator
`inhibitor type 1, its natural antagonist, to further regu-
`late endogenous thrombolysis.15
`There are complex interactions between the endothe-
`lium and thrombin, with thrombin promoting vasodila-
`tation where the endothelium is intact, through release
`of prostacyclin and nitric oxide, but causing vasocon-
`striction where endothelium is damaged, through liber-
`ation of endothelin.16 Thrombin also acts as an effector
`molecule, occupying receptors on the endothelium and
`smooth muscle cells to activate cellular proliferation,17
`as well as stimulate the release of platelet-derived
`growth factor (PDGF).18 Leukocyte and monocyte
`chemotaxis19 and release of interleukin-1 from macro-
`phages are also induced by thrombin.20
`The unique structure of thrombin is believed to
`account for its high specificity for its substrates. The
`location of the catalytic binding site in a deep narrow
`canyon on the thrombin molecule surface appears to
`restrict access to other macromolecules by steric hin-
`drance.21 Thrombin, distinct from most other serine
`
`proteases, has an anion-binding exosite, also known as
`the substrate recognition site, that is separate but
`adjacent to the active catalytic site.21 Fibrinogen binds
`at this site, increasing thrombin's specificity for it.
`Thrombomodulin and hirudin also attach here, whereas
`heparin appears to have its own basic binding site.2'
`Direct thrombin inhibitors known to block the anion-
`binding site have been shown to inactivate thrombin
`without displacing the thrombin molecule from the
`fibrin surface, providing evidence that thrombin has a
`further and separate binding site for fibrin distinct from
`its fibrinogen binding site.12
`The apolar binding site is located in a hydrophobic
`pocket close to the catalytic site in the fibrinopeptide
`groove22 and is also involved in substrate binding to the
`active catalytic site, as well as thrombin attachment to
`the platelet receptor glycoprotein lb. The main platelet
`receptor for thrombin, however, was recently character-
`ized by Coughlin and colleagues23 and has the unique
`feature of a tethered ligand. After thrombin binds the
`receptor through a hirudin-like anion-binding exosite, it
`cleaves the receptor to reveal a new receptor's amino
`acid terminus, which functions as the tethered ligand to
`activate the receptor. The molecule of thrombin thus
`remains free to activate other receptor sites and prop-
`agate the thrombotic process.23
`The regulation of thrombin is complex, involving
`proteinase inhibitors antithrombin III, heparin cofactor
`II, and a2-macroglobulin. Thrombin generation is con-
`trolled through coagulation amplification via factors
`including V and VIII; interaction with thrombomodulin,
`protein C, and protein S; as well as partitioning in the
`clot itself.24
`
`Direct Thrombin Inhibitors
`
`History
`The therapeutic use of the leech (Hirudo medicinalis)
`probably dates back to the ancient Greeks,25 but it was
`Haycraft who first described the antithrombotic prop-
`
`Downloaded from
`
`http://circ.ahajournals.org/
`
` by guest on October 7, 2015
`
`ANACOR EX. 2007 - 2/16
`
`

`
`1524
`
`Circulation
`
`Vol 90, No 3
`
`September 1994
`
`<}Cablyfc
`
`|
`
`Thrombin
`
`Thrombl n
`
`pro
`AN
`Unker
`
`Subtrate
`Recogntlon Site
`
`Thrombin
`
`Thrombn
`
`PPACK
`PA
`Argatban
`
`Schematic of the thrombin molecule, illustrating its key binding sites. Thrombin interactions with the direct thrombin inhibitors
`FIG 2.
`hirudin, hirulog, hirugen, PPACK, and argatroban are illustrated.
`
`erty of leech saliva in 1884.26 Hirudin was isolated by
`Markwardt in the 1950s,27 and the structure has subse-
`quently been characterized and cloned.28 Further devel-
`opment based on the structure of hirudin has led to the
`formation of a family of hirudin-like peptides called
`hirugens or hirullins.29 Peptide analogues hirugen and
`hirulog are based on the structure of hirudin, and
`synthetic derivatives argatroban, D-phenylalanyl-L-pro-
`lyl-L-arginyl chloromethylketone (PPACK), and the
`boroarginine derivatives have subsequently been devel-
`oped (Fig 2).
`Structure and Function
`Hirudin
`The prototypic thrombin inhibitor is the naturally
`occurring hirudin, isolated from the leech saliva, and is
`the most potent and specific known inhibitor of throm-
`bin, forming a tight, highly stable noncovalent complex
`with thrombin. It is a 65-amino-acid protein with three
`disulfide bridges and a molecular weight of -7000 d.
`Multiple isoforms of natural hirudin exist, with slightly
`different numbers of amino acids but similar anticoag-
`ulant activities. All have the disulfide bridges and a
`sulfated tyrosine residue in position 63.29
`Recombinant DNA technology has allowed the de-
`velopment of recombinant hirudin (r-hirudin), which
`has an identical amino acid sequence to natural hirudin,
`with or without the sulfated Tyr63.30 The nonsulfated
`molecule results in about a 10-fold reduction in throm-
`bin affinity,31 although binding still remains strong.
`Several companies have now developed proprietary
`versions of r-hirudin, which are at various stages of
`clinical testing. Although direct head-to-head compari-
`sons of these different recombinant have not been
`performed, Longstaff and colleagues,32 in a study de-
`signed to investigate standardization of laboratory hiru-
`din assays, found that four proprietary formulations of
`
`r-hirudin all had similar potencies, using a simple,
`standardized chromogenic assay.
`The hirudin molecule has two distinct domains-an
`NH2 terminal core domain and a COOH terminal tail.27
`The N-terminal binds and inhibits the active catalytic
`site of thrombin,27 whereas the carboxy terminal simul-
`taneously blocks the anion-binding exosite33 (Fig 2).
`This interaction takes place in two steps, with an initial
`ionic interaction and then a rearrangement of the
`hirudin-thrombin complex to form a stronger binding.34
`The apolar-binding site also appears to be involved in
`hirudin binding, since the compound proflavin (which
`binds to the apolar-binding site) is displaced during
`formation of the hirudin-thrombin complex.22 Crystal
`structure studies have further demonstrated that the
`attachment of hirudin to thrombin is not limited to the
`three binding sites but also forms multiple other con-
`tacts, tightly binding thrombin over an extended area of
`the molecule.3536 Overall, this allows hirudin to be
`uniquely specific for thrombin, with no inhibitory action
`on any other serine protease.36
`Hirugen
`Hirugen, a synthetic hirudin derivative, is a dode-
`capeptide comprising the terminal 12 residues of hiru-
`din involved in anion-binding exosite blockade (Fig 2).
`The binding of hirugen to thrombin is similar to the
`C-terminal of hirudin,37 and the molecule contains a
`sulfated tyrosine to increase its thrombin afiinity. Hir-
`ugen was designed to block the interaction of thrombin
`with fibrinogen, leaving the catalytic site of thrombin
`free to interact with antithrombin III, enhancing its
`actions, with or without heparin.38 However, the in vivo
`antithrombotic activity of hirugen is considerably
`weaker compared with hirudin and hirulog,39 and it has
`not progressed to testing in the clinical arena.
`
`Downloaded from
`
`http://circ.ahajournals.org/
`
` by guest on October 7, 2015
`
`ANACOR EX. 2007 - 3/16
`
`

`
`Hirulog
`A group of peptides called hirulogs were subse-
`quently developed that had the D-Phe-Pro-Arg se-
`quence of the N-terminal of hirudin linked to the
`hirugen molecule by a segment of glycyl residues varying
`from 6 to 18 A in length.40 Like hirudin, this synthetic
`derivative is able to block both the active catalytic site
`and anion-binding site of the thrombin molecule (Fig 2).
`Hirulog-1 has four glycyl residues between the two
`domains and has been found to be a potent thrombin
`inhibitor.40 Although hirulog retains the high specificity
`of hirudin for thrombin, its substrate-type binding to
`thrombin is more typical of other protease inhibitors, in
`contrast to the unique multiple contact binding of
`hirudin.36 However, doubt has arisen regarding the
`contribution of active site inhibition to hirulog's action,
`following the discovery that thrombin itself mediates
`slow cleavage of hirulog at the Arg-Pro bond on the
`amino-terminal extension in vivo,41 effectively convert-
`ing it into a hirugen-like molecule. This may account for
`some of the pharmacological differences with hirudin40
`and has led to the development of hirudin derivatives
`with noncleavable bonds.42
`PPACK and Argatroban
`PPACK is a tripeptide synthetic compound that has a
`structure very close to fibrinopeptide A.43 It contains
`the amino acid sequence corresponding to the cleavage
`site of fibrinogen and acts as an affinity agent to
`thrombin.44 PPACK blocks the active catalytic site by
`alkylating the active site histidine residue and is an
`irreversible inhibitor of thrombin (Fig 2). The com-
`pound has been found to be an effective thrombin
`inhibitor in plasma and in vivo, but the presence of a
`reactive group leads to a relatively rapid loss of activity
`due to reactions with other plasma components.45 Ar-
`gatroban, an arginine derivative, binds thrombin at the
`apolar-binding site adjacent to the active site. Its main
`action is to block thrombin's active catalytic site as does
`PPACK, but unlike this compound, argatroban is a
`competitive antagonist. Comparative studies with other
`thrombin inhibitors have shown this compound to be a
`potent thrombin inhibitor with strong affinity for
`thrombin.44,46
`Other Thrombin Inhibitors
`Many other compounds have been developed that
`have antithrombin activity, but are too toxic for clinical
`use (Table 1). Thrombin inhibition by arginine and
`benzamidine derivatives ranges from weak to strong,44"45
`whereas the boroarginine derivatives such as DuP 714
`are potent antithrombins and have the potential for oral
`bioavailability.47 However, to date use has been limited
`by adverse liver toxicity, believed to be related to the
`boron constituent.
`Several other classes of antithrombins, at various
`stages of development, are aimed at controlling the
`thrombotic process at different strategic points or
`through novel methods. The low molecular weight hep-
`arin (LMWH) preparations are fragments of standard
`heparin obtained by chemical or enzymatic depolymer-
`ization of the polysaccharide chains. Although they are
`antithrombin III dependent and cannot inhibit throm-
`bin that is fibrin bound, they are differentiated from
`
`Lefkovits and Topol
`
`Direct Thrombin Inhibitors
`
`1525
`
`Classification of Antithrombotic Agents
`TABLE 1.
`Antithrombin Ill dependent
`Heparin
`Low molecular weight heparins
`Heparinoids
`Direct thrombin inhibitors
`Hirudin
`Hirugen
`Hirulog
`Argatroban
`PPACK
`DuP 714
`Thrombin aptamers
`Thrombin generation inhibitors
`Factor Xa inhibitors
`Antistasin
`Tick anticoagulant peptide
`Inactivated factor X
`Factor VIl antibody and peptidomimetics
`Tissue factor pathway inhibitor
`Recombinant endogenous anticoagulants
`Activated protein C
`Antithrombin Ill
`Heparin cofactor 11
`Tissue factor pathway inhibitor
`Thrombin receptor blockers
`Thrombin receptor antagonist peptides
`Vitamin K antagonists
`Warfarin sodium
`
`heparin by both their relative and absolute increased
`affinity for factor Xa over thrombin,48 with the degree of
`selectivity for factor Xa varying with different commer-
`cial preparations. The LMWH compounds have lower
`affinities for endothelial cells and plasma proteins than
`standard heparin and are therefore cleared more slowly,
`allowing once-daily dosing and more predictable anti-
`coagulant responses.48 Several trials have now demon-
`strated their clinical use, particularly in the areas of
`deep venous thrombosis prophylaxis and treatment.49
`A novel mechanism for control of the thrombotic
`process has been developed using single-stranded DNA
`oligonucleotides known as aptamers to bind target
`proteins and block their actions. Thrombin aptamers,
`first described by Bock and colleagues,50 bind the fibrin-
`ogen recognition site on the thrombin molecule51 and
`have demonstrated potent antithrombotic effects in
`vitro,50 in animal models of thrombosis,5253 and also
`have the ability to inhibit clot-bound thrombin.53 The 15
`nucleotide aptamer used in a number of studies has the
`distinct characteristics of rapid onset of action and short
`half-life, which may favor its future use in certain acute
`clinical settings.52'53
`Alternative strategies for controlling thrombin activ-
`ity are becoming available through the isolation and
`
`Downloaded from
`
`http://circ.ahajournals.org/
`
` by guest on October 7, 2015
`
`ANACOR EX. 2007 - 4/16
`
`

`
`1526
`
`Circulation
`
`Vol 90, No 3
`
`September 1994
`
`development of agents that selectively inhibit key en-
`zymes at different points within the coagulation cas-
`cade. From a theoretical standpoint, because the direct
`thrombin inhibitors antagonize thrombin that is pre-
`formed, some thrombin activity may still occur before its
`neutralization. New formation of thrombin remains
`unaffected and may also contribute to a "leakage" of
`thrombin activity despite the presence of an inhibitor.
`Conversely, inhibition of activated factor X (factor Xa)
`can prevent new formation of thrombin and disrupt the
`thrombin feedback loop that autoamplifies thrombin
`generation but has no effect on thrombin that is already
`present. Inhibition of proteins earlier in the pathway,
`such as tissue factor - a key initiator of thrombosis in
`vivo - offers a more mechanistically specific targeting
`for prevention of thrombosis but may have limited
`efficacy because of compensatory augmentation in ac-
`tivity of alternate coagulation pathways such as the
`intrinsic pathway.
`With these theoretical considerations in mind, a
`number of thrombin generation inhibitors have under-
`gone early experimental testing. Local inhibition of
`factor Xa has been achieved by antistasin, a 119-amino-
`acid protein initially isolated from the salivary gland of
`the Mexican leech,54 and tick anticoagulant peptide
`(TAP) derived from the soft tick,55 now produced by
`recombinant technology. Both are potent and highly
`selective antagonists of factor Xa,55,56 and unlike anti-
`thrombin III, can inhibit factor Xa attached to the
`prothrombinase complex as well as free factor Xa.56
`When used as adjuncts to thrombolysis, these agents,
`through the inhibition of thrombin generation, acceler-
`ated clot lysis and prevented reocclusion.5758 This has
`brought into focus the controversial concept that, at
`least in part, it is the continuous generation of new
`thrombin rather than reexposure of preformed clot-
`bound thrombin that is responsible for the phenomenon
`of reocclusion. Indeed, markers of thrombin generation
`such as prothrombin fragment F1.2 have been found to
`increase during and after thrombolytic treatment for
`acute myocardial infarction, suggesting that increased
`thrombin activity associated with thrombolysis is, at
`least in part, due to new thrombin generation.59 This in
`turn may have important implications for the eventual
`place of direct thrombin inhibitors as adjuncts to throm-
`bolysis, as lack of inhibition of new thrombin generation
`represents a potential mechanistic limitation of these
`agents.
`Recombinant forms of naturally occurring endoge-
`nous anticoagulants is yet another strategy that has
`been exploited to control the thrombotic process. Tissue
`factor pathway inhibitor (TFPI) is a multivalent protein-
`ase inhibitor, whose complex actions involve the binding
`and inhibition of factor Xa, producing a factor Xa-TFPI
`complex that causes feedback inhibition on the factor
`VIIa-tissue factor complex responsible for triggering
`factor X activation.6061 Purified and recombinant acti-
`vated Protein C, through their inhibition of factors Va
`and VIIIa, inhibit thrombin formation62 and have also
`been found to inactivate plasminogen activator inhibi-
`tor.63 In vitro and animal studies have demonstrated
`antithrombotic effects of this agent,63,64 and human
`evaluation in phase I trials has already commenced.65 A
`potential use for this compound is as an adjunct to
`thrombolysis, particularly following the discovery that
`
`thrombolytic therapy itself appears to generate acti-
`vated protein C.66 Novel approaches such as monoclonal
`antibodies against or peptidomimetics of factor VII and
`antibodies against the recently characterized platelet
`thrombin receptor are all in the very early stages of
`development.
`Experimental Studies With Direct
`Thrombin Inhibitors
`As expected from their mode of action, direct throm-
`bin inhibitors antagonize the broad spectrum of throm-
`bin's actions. Hirudin has been shown to inhibit fibrin
`formation; prevent activation of factors V, VIII, and
`XIII; and prevent thrombin-mediated platelet activa-
`tion in vitro.67,68 It produces virtual total inhibition of
`thrombus formation ex vivo in both rat and human
`plasma30 and may also augment displacement of factor
`Xa from vascular endothelium, actually associating it-
`self with factor Xa.69 Animal studies have confirmed the
`antithrombotic efficacy of hirudin in a wide range of
`animal venous and arterial models of thrombo-
`sisj27,30,70-73 Hirulog,39,74 argatroban,75,76 and PPACK77
`have also proven to be effective antithrombotics in
`various animal models. Hirudin appears to have addi-
`tional effects on thrombus composed predominantly of
`platelets rather than fibrin,78 although in a rat dissemi-
`nated intravascular coagulation model higher concen-
`trations of hirudin were required to inhibit platelet
`deposition than to prevent fibrin deposition.79
`Direct thrombin inhibitors may have an especially
`important adjunctive role in thrombolysis. Apart from
`their main action to activate plasmin and lyse fibrin,
`thrombolytics also trigger a number of reactions that
`engender a paradoxically "prothrombotic state," includ-
`ing plasmin-mediated activation of factor XII and plate-
`let activation,80 platelet release of plasminogen activator
`inhibitor,81 and consumption of plasma plasminogen,
`which in turn draws clot-bound plasminogen into the
`soluble phase, reducing the substrate for activation and
`attenuating fibrinolytic efficacy (plasminogen steal).82
`Of chief significance, however, is that lytic agents accel-
`erate exposure of thrombin bound to fibrin within the
`clot, as well as increase the amount of free throm-
`bin,83-85 producing a milieu highly conducive to re-
`thrombosis. Experimental models both in vitro and in
`vivo have demonstrated the ability of thrombin inhibi-
`tors to neutralize both clot-bound thrombin and free
`soluble thrombin.1286-88 Hirudin has been shown to
`accelerate thrombolysis following TPA,89 and prevent
`reocclusion in a canine model,90 whereas hirulog accel-
`erated thrombolysis and prevented reocclusion in a rat
`model.74 Argatroban, an agent with less affinity for
`thrombin than either hirudin or hirulog, was able to
`accelerate clot lysis by TPA in a canine stenotic coro-
`nary artery model but required the addition of aspirin
`to reduce the incidence of reocclusion.9'
`There is further compelling experimental evidence to
`support a beneficial role for direct thrombin inhibitors
`in the field of coronary intervention. Platelet aggrega-
`tion and fibrin deposition produced in a pig model of
`carotid angioplasty have been successfully prevented by
`hirudin,78 and the same agent also inhibited restenosis
`at 28 days in a rabbit angioplasty model, suggesting an
`additional mechanism of inhibition of thrombin's effec-
`tor molecule function as a stimulator of smooth muscle
`
`Downloaded from
`
`http://circ.ahajournals.org/
`
` by guest on October 7, 2015
`
`ANACOR EX. 2007 - 5/16
`
`

`
`proliferation.92 Thrombin inhibitors effectively pre-
`vented thrombin-induced platelet activation,93,94 and
`comparative studies with heparin found hirudin to be
`more effective in reducing both platelet and fibrin
`deposition on coronary stents,95 platelet thrombi in the
`porcine deep arterial injury model,78 and more potently
`inhibited thrombus in rabbit and rat thrombosis
`models.88,96
`Clinical Pharmacology and Phase I Studies
`Relatively little data are available from comparative
`and systematic pharmacological trials, although hirudin
`remains the most extensively studied thrombin inhibitor
`pharmacologically. Recombinant hirudin is poorly ab-
`sorbed orally and is administered parenterally via the
`intravenous, intramuscular, or subcutaneous route. Pri-
`mate studies have demonstrated faster onset and earlier
`return to baseline after intravenous administration. The
`half-life after intravenous use was 40 minutes compared
`with almost 2 hours after subcutaneous injection.70 The
`volume of distribution has variously been estimated
`from 8.9 to 17.2 L97 with significant distribution into the
`extravascular space. Little hepatic metabolism takes
`place with hirudin, and 95% of the compound is renally
`excreted in its active form.70'97 Impaired elimination has
`been demonstrated in patients with renal insufficiency,97
`and dose reduction is recommended in patients with
`renal dysfunction.
`In a recent pharmacological trial with human volun-
`teers,98 hirulog differed from hirudin in that its half-life
`was 15 to 20 minutes shorter after intravenous bolus
`administration and that total body clearance rate was
`more rapid at 0.5 L/min for a 70-kg man compared with
`hirudin's clearance rate of 0.19 L/min.99 However, only
`20% of hirulog was recovered from the urine, suggesting
`that it undergoes more extensive metabolic clearance
`than hirudin, including hepatic metabolism or proteol-
`ysis at other sites.
`Several phase I trials have now been completed with
`the various thrombin inhibitors, demonstrating excel-
`lent safety and anticoagulant efficacy profiles in healthy
`volunteers. Argatroban was well tolerated in healthy
`human volunteers and produced a dose-dependent pro-
`longation of the coagulation parameters activated par-
`tial thromboplastin time (aPTT) and thrombin time
`(TT), with a return to baseline levels 1 hour after
`intravenous administration. No prolongation of skin
`bleeding time was detected, even with concomitant
`administration of aspirin.100
`The biologic effects of sulfated r-hirudin (CGP-39393)
`were comprehensively studied in a recent report by
`Verstraete and colleagues.10' Hirudin was administered
`subcutaneously, as either a single dose or repeated doses,
`in 231 healthy volunteers. The drug was extremely well
`tolerated, with no significant adverse events reported.
`Single injections of 0.1 mg/kg prolonged the aPTT to
`almost twice baseline values in 184 volunteers, and
`increasing doses up to 0.75 mg/kg increased the aPTT
`linearly. The anticoagulant effect after subcutaneous
`administration was noted within 30 minutes, peaking at 4
`to 6 hours, and returning to baseline within 24 hours.
`Similar findings have been reported elsewhere.99102"103
`Repeated subcutaneous doses lengthened aPTT for as
`long as the administration was continued, with no evi-
`dence of cumulative effects even up to 6 days of admin-
`
`Lefkovits and Topol
`
`Direct Thrombin Inhibitors
`
`1527
`
`istration. Bleeding time was not prolonged even up to
`doses of 0.5 mg/kg, suggesting little effect of hirudin on
`primary hemostasis.10'
`Tolerability and anticoagulant activity of hirulog
`were assessed in a recent randomized, placebo-con-
`trolled study of 54 healthy volunteers,98 with both
`intravenous and subcutaneous routes of administration
`studied. Doses ranged from 0.05 to 1.0 mg/kg, and a
`rapid dose-dependent prolongation of the aPTT for
`both intravenous and subcutaneous routes was demon-
`strated. There was no significant bleeding or other
`adverse events reported, and overall there was no
`prolongation of the skin bleeding time. Subcutaneous
`injection resulted in a later peaking of anticoagulant
`effect and a sustained prolongation of aPTT over sev-
`eral hours. Greater anticoagulant activity was achieved
`at lower doses of hirulog by extending the intravenous
`infusion period for 2 hours or more, after steady-state
`levels of the drug were obtained.98
`Clinical (Phase II) Studies
`Several studies have assessed the role of thrombin
`inhibitors in acute ischemic syndromes and PTCA
`(Table 2).
`Acute Myocardial Infarction
`A number of small studies published recently have
`assessed thrombin inhibitors as adjuncts to thromboly-
`sis. Lidon et al'04 randomized 42 patients to heparin or
`hirulog after streptokinase for acute myocardial infarc-
`tion. Hirulog had significantly better 90-minute Throm-
`bolysis in Myocardial Infarction (TIMI)-3 patency rates
`(61% versus 29%, P=.05) and 120-minute patency rates
`(75% versus 36%, P<.02). Bleeding complications were
`the same in the two groups. Tabata et al105 evaluated
`argatroban in the prevention of reocclusion in 24 pa-
`tients following successful reperfusion therapy for myo-
`cardial infarction. Reocclusion at 1 month was 0%
`compared with 15% in 74 patients on heparin control.
`This effect was independent of the type of reperfusion
`therapy or residual stenosis after treatment.
`Results of the Hirudin for the Improvement of
`Thrombolysis (HIT) study were recently published.106
`In this dose-escalation study, three doses of recombi-
`nant hirudin (HBW 023) were used as adjunctive ther-
`apy following front-loaded TPA for myocardial infarc-
`tion in 143 patients. Patency rates were determined at
`30, 60, and 90 minutes; at 36 to 48 hours; and after
`discharge. Ninety-minute TIMI-3 patency rates in-
`creased, and reocclusion rates diminished with increas-
`ing doses of hirudin. This trial did not have a heparin
`control arm, but patency rates compared favorably with
`those previously established for heparin, and the reoc-
`clusion rates were lower than the 10% to 15% rate
`associated with adjunctive heparin therapy. There were
`three spontaneous hemorrhages in the entire study
`group, and an increase in puncture site bleeding was
`noted in the highest dose group (5 of 83 patients).
`Cannon and colleagues107 reported the findings of the
`TIMI-5 trial that randomized 246 patients to either
`hirudin or heparin for 5 days following