Anticoagulants, Antiplatelets, and Thrombolytics: Second Edition

METHODS IN MOLECULAR BIOLOGYTM

Series EditorJohn M. Walker

School of Life SciencesUniversity of Hertfordshire

Hatfield, Hertfordshire, AL10 9AB, UK

For other titles published in this series, go towww.springer.com/series/7651

Anticoagulants, Antiplatelets,and Thrombolytics

Second Edition

Edited by

Shaker A. MousaPharmaceutical Research Institute, Albany College of Pharmacy

and Health Sciences, Rensselaer, NY, USA

EditorShaker A. MousaAlbany College of Pharmacy

and Health SciencesPharmaceutical Research InstituteOne Discovery DriveRensselaer, NY [email protected]

ISSN 1064-3745 e-ISSN 1940-6029ISBN 978-1-60761-802-7 e-ISBN 978-1-60761-803-4DOI 10.1007/978-1-60761-803-4Springer New York Dordrecht Heidelberg London

Library of Congress Control Number: 2010929846

© Springer Science+Business Media, LLC 2003, 2010All rights reserved. This work may not be translated or copied in whole or in part without the written permission ofthe publisher (Humana Press, c/o Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013,USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form ofinformation storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodologynow known or hereafter developed is forbidden.The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identifiedas such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights.While the advice and information in this book are believed to be true and accurate at the date of going to press, neitherthe authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that maybe made. The publisher makes no warranty, express or implied, with respect to the material contained herein.

Printed on acid-free paper

Humana Press is part of Springer Science+Business Media (www.springer.com)

Preface

The past 2 decades have witnessed significant advances in the discovery and developmentof novel drugs to prevent and treat thromboembolic disorders, such as oral direct anti-Xa and anti-IIa (thrombin) antagonists, as well as oral antiplatelet ADP antagonists withrapid onset and offset. The introduction of direct oral factor Xa and thrombin inhibitorsthat do not require monitoring and have no significant food or drug interactions repre-sents a significant advance that may lead to the replacement of oral warfarin and injectableheparin or low molecular weight heparin (LMWH) in some, but probably not all, indi-cations. In addition, there has been concentrated effort aimed at identifying novel usesof traditional antithrombotic drugs as well as combinations of agents, such as more thanone antiplatelet, or antiplatelet plus anticoagulant. These tremendous achievements haveresulted in improved management of arterial and venous thromboembolic-associated dis-orders. Although the morbidity and mortality resulting from acute coronary disease hasbeen reduced by more than 50% over the past 30 years, it is reasonable to anticipate furtherreductions of similar magnitude in the decade ahead.

Advances in our understanding of the mechanisms of pathogenesis of venous throm-boembolism (VTE), acute coronary syndromes, cerebral vascular ischemia, and diseasesassociated with thrombotic events have provided critical insight for the development ofvarious therapeutic approaches to control these pathogenic events. The roles of plas-matic proteins, blood cells, vascular endothelium, and target organs in thrombogene-sis are becoming more clear. Identification of endogenous inhibitors of thrombogenesissuch as antithrombin III, tissue factor pathway inhibitor (TFPI), protein C, prostacyclin,nitric oxide, and physiologic activators of fibrinolysis has led to the development of bothdirect and indirect modalities to treat thrombosis. Knowledge of the proteases involved inthrombogenesis, as well as tissue factor, coagulation factors, adhesion molecules, and fib-rinolytic inhibitors, has provided additional insight into the mechanisms by which throm-bogenesis can be pharmacologically controlled. All of these novel strategies could not havehappened without the utilization of key in vitro and in vivo clinically relevant experimentalmodels for the screening and evaluation of these novel antiplatelets, anticoagulants, andthrombolytics (discussed in Chapters 1 and 2).

Newly developed anti-Xa agents are characterized by high affinity and selectivity for Xaas compared to other serine proteases. In addition to their inhibitory effects on plasmaticcoagulation processes, including thrombin generation, thrombin-mediated platelet reac-tions, and clot-bound pro-thrombinase complexes, there is evidence that some of theseagents might interfere with receptor-mediated intracellular signaling events induced byfactor Xa that regulate proliferation of vascular smooth muscle cells and other cells. Thecurrent outlook for anti-Xa agents is that they have the potential to become importantprophylactic and treatment drugs for various venous thromboembolic disorders as well asadjuvants to other antithrombotic therapies in arterial thrombosis.

Major advances in the development of oral anticoagulants are progressing very well,with the goal of developing safe and effective oral anticoagulants that do not requirefrequent monitoring or dose adjustment and that have minimal food/drug interactions.

v

vi Preface

Vitamin K antagonist, with its inherent limitations of multiple food and drug interactionsand frequent need for monitoring, remains the only oral anticoagulant currently approvedfor long-term secondary thromboprophylaxis in VTE. The oral direct thrombin inhibitorximelagatran was withdrawn from the world market due to safety concerns. Newer anti-coagulant drugs such as injectable pentasaccharides (e.g., idraparinux, SSR126517E), oraldirect thrombin inhibitors (e.g., dabigatran), oral direct factor Xa inhibitors (e.g., rivarox-aban, apixaban, YM-150, DU-176b), and tissue factor/factor VIIa complex inhibitors are“tailor-made” to target specific pro-coagulant complexes and have the potential to greatlyexpand oral antithrombotic targets for both acute and long-term treatment of VTE, acutecoronary syndromes, and prevention of stroke in atrial fibrillation patients (discussed inChapter 5).

The oral direct factor Xa inhibitor rivaroxaban represents a potentially attractive alter-native to warfarin as it may enable simplified once daily dosing, appears to require no ther-apeutic monitoring, and has lower potential for drug interactions. At present, the safetyand efficacy of rivaroxaban for the prophylaxis and treatment of VTE have been evaluatedin phase-II and phase-III trials involving over 24,000 patients. In addition, rivaroxabanis currently being evaluated for the treatment of pulmonary embolism, secondary preven-tion after acute coronary syndromes, and the prevention of stroke and non-central nervoussystem embolism in patients with non-valvular atrial fibrillation. Several other oral directanti-Xa inhibitors are in advanced clinical development, approved in Europe and underFDA review for the prevention and treatment of thromboembolic disorders (discussed inChapter 6).

Dabigatran is a novel oral direct reversible (fast onset and offset) thrombin inhibitorthat binds to both free and clot-bound thrombin with a high affinity and specificity.Dabigatran has predictable and reproducible pharmacokinetics that are not affected byinteractions with food. It is not metabolized by CYP450, does not induce nor inhibitCYP450, resulting in low potential for drug interactions, and does not require coagulationor platelet monitoring. The RE-NOVATE trial demonstrated that oral dabigatran etexi-late at fixed doses is a well-tolerated alternative to injectable enoxaparin for the preventionof VTE after total knee replacement. The RELY trial demonstrated that oral dabigatranetexilate concurrently reduces both thrombotic and hemorrhagic events at two differentdoses (150 and 110 mg BID), exhibiting different and complimentary advantages overwarfarin. At a dose of 150 mg BID, dabigatran had superior efficacy with similar bleeding,while at a dose of 110 mg BID, there was significantly less bleeding with similar efficacyin patients with atrial fibrillation at risk of stroke. Based on the accumulating clinical evi-dence, dabigatran represents the future of anticoagulation in the prevention and treatmentof venous and arterial thrombosis alone and in conjunction with current antiplatelets andthrombolytics.

Anti-platelet therapies remain a major focus in drug development. While aspirin is stillconsidered the gold standard for antiplatelet therapy because of its high benefit-to-costand benefit-to-risk ratios, ADP receptor antagonists, including ticlopidine, clopidogrel,and prasugrel, represent significant additions to aspirin in the management of differentforms of arterial thromboembolic disorders (Chapter 7). Prasugrel is a novel thienopy-ridine that inhibits the platelet P2Y12 receptor and provides more rapid and consistentplatelet inhibition than clopidogrel (Chapter 8).

It is becoming clear, however, that there is variability in individual responses toantiplatelet agents such as clopidogrel, which may limit their widespread implementa-tion. Various definitions of “non-responders” to antiplatelet therapy (i.e., aspirin resis-

Preface vii

tance) tend to compound this issue. Aspirin resistance refers to aspirin-treated patientsthat are insensitive to aspirin treatment based on ex vivo tests of platelet activation andwho experience recurrent cardiovascular disease. Estimates of aspirin resistance based onthese criteria range from 20 to 80%, indicating that ex vivo tests are not an optimal toolfor such assessments. In long-term aspirin-treated patients, there is evidence of low levelbut functionally relevant platelet thromboxane A2 formation, which was responsible forenhanced platelet activation in response to platelet agonists. These studies, however, didnot fully exclude aspirin compliance, which could be a factor in such a phenomenon. Twotrials performed in patients with coronary artery disease demonstrated that laboratoryevidence of aspirin resistance was not detectable when aspirin compliance was accuratelymonitored. The same phenomenon was reported for other anti-platelet drugs such asclopidogrel. Given the multi-factorial nature of atherothrombosis, recurrence of cardio-vascular events in aspirin-treated patients is not necessarily suggestive of drug failure. Acause-effect relationship between platelet insensitivity to aspirin and cardiovascular recur-rence has not been defined overall because aspirin compliance has rarely been considered.Until such crucial information is taken into account, it would be prudent to take into con-sideration the distinction between “clinical resistance” to aspirin and resistance to takingthe drug. To carefully define anti-platelet resistance, issues such as dose levels, standard-ized monitoring parameters, drug-drug interactions, and drug monitoring to documentcompliance should be addressed in future studies.

The next decade should see considerable attention focused on the vascular endothe-lium, which occupies a strategic position at the interface between tissue and blood. Thenormal endothelium releases multiple antiplatelet, anti-inflammatory, thrombolytic andvasodilator molecules, such as prostacyclin and nitric oxide, which are potent inhibitorsof platelet and monocyte activation and function as vasodilators. In addition, the nor-mal endothelial surface expresses other protective molecules, including ecto-ADP, whichdegrades ADP, leading to inhibition of platelet aggregation; thrombomodulin, whichactivates protein C; and heparin-like molecules, which serve as cofactors for antithrom-bin III and heparin. The normal endothelium also secretes tissue plasminogen activator,which activates fibrinolysis. Insult or injury to the endothelium is accompanied by lossof these protective molecules and induction of expression of adhesive, pro-coagulant andpro-inflammatory molecules, vasoconstrictors, and mitogenic factors, leading to the devel-opment of thrombosis, smooth muscle cell migration and proliferation, and atherosclero-sis. Hence, protective mechanisms of endothelial function represent new frontiers in theprevention and treatment of thromboembolic disorders that will have minimal effect onhemostasis.

Improved understanding of the cell biology of plaque instability and endothelialhemostasis will promote a number of novel therapeutic strategies, including passivation ofthe endothelium, reduction of low-density lipoprotein (LDL) in the vessel wall (throughdecreasing serum LDL levels or accelerating reverse cholesterol transport), inhibition ofLDL oxidation, thereby raising high density lipoprotein (HDL), and inhibition of inflam-matory cytokine expression, as well as inhibition of thrombus formation upstream in thecoagulation cascade or inhibition of activation of coagulation. The recognition that throm-botic disorders represent a syndrome rather than a disease is of crucial importance in thedevelopment of newer drugs. Either a poly-therapeutic approach with drug combinationsor a drug with multiple actions will likely be more appropriate for the management ofthrombotic disorders.

viii Preface

This second edition of Anticoagulants, Antiplatelets, and Thrombolytics providesupdates on various strategies in thrombosis, experimental models, and clinical and recentadvances in the discovery and development of novel antithrombotics. Future directions inthe coming decade should focus on the prevention of thromboembolic disorders and theprotection of the vascular endothelium.

Shaker A. Mousa

ContentsPreface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v

Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi

1. In Vitro Methods of Evaluating Antithrombotics and Thrombolytics . . . . . . . 1Shaker A. Mousa

2. In Vivo Models for the Evaluation of Antithrombotics and Thrombolytics . . . . 29Shaker A. Mousa

3. Heparin and Low-Molecular Weight Heparins in Thrombosis and Beyond . . . . 109Shaker A. Mousa

4. Laboratory Methods and Management of Patients with Heparin-InducedThrombocytopenia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133Margaret Prechel, Walter P. Jeske, and Jeanine M. Walenga

5. Novel Anticoagulant Therapy: Principle and Practice . . . . . . . . . . . . . . . 157Shaker A. Mousa

6. Oral Direct Factor Xa Inhibitors, with Special Emphasis on Rivaroxaban . . . . . 181Shaker A. Mousa

7. Antiplatelet Therapies: Drug Interactions in the Managementof Vascular Disorders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203Shaker A. Mousa

8. Prasugrel: A Novel Platelet ADP P2Y12 Receptor Antagonist . . . . . . . . . . . 221Shaker A. Mousa, Walter P. Jeske, and Jawed Fareed

9. Antithrombotic Effects of Naturally Derived Products on Coagulationand Platelet Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229Shaker A. Mousa

10. Assessment of Anti-Metastatic Effects of Anticoagulant and AntiplateletAgents Using Animal Models of Experimental Lung Metastasis . . . . . . . . . 241Ali Amirkhosravi, Shaker A. Mousa, Mildred Amaya, Todd Meyer,Monica Davila, Theresa Robson, and John L. Francis

11. Adhesion Molecules: Potential Therapeutic and Diagnostic Implications . . . . . 261Shaker A. Mousa

12. Pharmacogenomics in Thrombosis . . . . . . . . . . . . . . . . . . . . . . . . 277Shaker A. Mousa

13. Diagnosis and Management of Sickle Cell Disorders . . . . . . . . . . . . . . . 291Shaker A. Mousa and Mohamad H. Qari

Subject Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309

ix

Contributors

MILDRED AMAYA • Florida Hospital Center for Thrombosis Research, Orlando, FL, USAALI AMIRKHOSRAVI • Florida Hospital Center for Thrombosis Research, Orlando,

FL, USAMAMDOUH BAKHOS • Department of Thoracic and Cardiovascular Surgery, Stritch

School of Medicine, Loyola University, Maywood, IL, USAMONICA DAVILA • Florida Hospital Center for Thrombosis Research, Orlando, FL, USAJAWED FAREED • Department of Pathology, Stritch School of Medicine, Loyola University,

Maywood, IL, USAJOHN L. FRANCIS • Florida Hospital Center for Thrombosis Research, Orlando, FL, USAWALTER P. JESKE • Department of Thoracic & Cardiovascular Surgery, Stritch School of

Medicine, Loyola University, Maywood, IL, USATODD MEYER • Florida Hospital Center for Thrombosis Research, Orlando, FL, USASHAKER A. MOUSA • Pharmaceutical Research Institute, Albany College of Pharmacy and

Health Sciences, Rensselaer, NY, USAMARGARET PRECHEL • Department of Pathology, Stritch School of Medicine, Loyola

University, Maywood, IL, USAMOHAMAD H. QARI • College of Medicine, King Abdul-Aziz University, Jeddah, Saudi

ArabiaTHERESA ROBSON • Florida Hospital Center for Thrombosis Research, Orlando, FL, USAJEANINE M. WALENGA • Department of Thoracic & Cardiovascular Surgery, Stritch

School of Medicine, Loyola University, Maywood, IL, USA

xi

Chapter 1

In Vitro Methods of Evaluating Antithromboticsand Thrombolytics

Shaker A. Mousa

Abstract

Platelets play a crucial role in primary hemostasis by forming hemostatic plugs at sites of vascular injury.There is abundant evidence that platelets also play a pivotal role in the pathogenesis of arterial throm-botic disorders, including unstable angina (UA), myocardial infarction (MI), and stroke. The underlyingpathophysiological mechanism of these processes has been recognized as the disruption or erosion of avulnerable atherosclerotic plaque, leading to local platelet adhesion and subsequent formation of partiallyor completely occlusive platelet thrombi. A variety of methods have been used to assess platelet aggrega-tion, blood coagulation, and the ex vivo and/or in vitro efficacy of platelet antagonists, anticoagulants,and thrombolytics.

Key words: Platelets, aggregation, coagulation, thrombosis, in vitro models, adhesion.

1. Introduction

The specific platelet surface receptors that support the initialadhesive interactions that ultimately lead to the formation ofthrombi are determined by the local fluid dynamics of the vas-culature and the extracellular matrix constituents exposed at thesites of vascular injury. Under high shear conditions, the adhesionof un-activated platelets to exposed sub-endothelial surfaces ofatherosclerotic or injured vessels is mediated by binding of plateletglycoprotein (GP) Ib/IX/V complex to collagen and von Wille-brand factor (vWF) presented on exposed vessel surfaces (1, 2).This primary adhesion to the matrix activates platelets, ultimatelyresulting in platelet aggregation, which is mediated predomi-nantly by the binding of adhesive proteins, such as fibrinogen

S.A. Mousa (ed.), Anticoagulants, Antiplatelets, and Thrombolytics, Methods in Molecular Biology 663,DOI 10.1007/978-1-60761-803-4_1, © Springer Science+Business Media, LLC 2003, 2010

1

2 Mousa

and vWF, to GPIIb/IIIa. Direct platelet aggregation in the bulkphase under conditions of abnormally elevated fluid shear stress,analogous to what occurs in atherosclerotic or constricted arterialvessels, may also be important (3). Shear-induced platelet aggre-gation is dependent upon the availability of vWF and the pres-ence of GPIb/IX and GPIIb/IIIa on the platelet membrane. Ithas been postulated that under high shear stress conditions, theinteraction of vWF with the GPIb/IX complex is the initial eventleading to platelet activation and also triggers the binding of vWFto GPIIb/IIIa to induce platelet aggregate formation.

The coagulation cascade consists of a complex network ofinteractions resulting in thrombin-mediated conversion of fib-rinogen to fibrin, a major component of the thrombus. Initiationof the coagulation cascade occurs through the release of throm-boplastin (tissue factor) and subsequent activation and conver-sion of factor VII into tissue factor/factor VIIa complex (“exoge-nous pathway”), or by the so-called contact activation pathway or“endogenous pathway,” which proceeds through factors XII, XI,and IX to the assembly of a tenase complex, consisting of acti-vated factors VIII and IX and Ca2+, on phospholipid membranes.Both the exogenous and endogenous tenase complex can acti-vate factor X, which induces the formation of the prothrombinasecomplex, consisting of factor Xa, factor Va, and Ca2+, on phos-pholipid surfaces. Assembly of the prothrombinase complex leadsto the activation of thrombin, which cleaves fibrinogen to yieldfibrin.

2. In VitroCoagulation Tests

2.1. BloodCoagulation Tests

2.1.1. Purposeand Rational

Three coagulation tests, prothrombin time (PT), activated partialthromboplastin time (aPTT), and thrombin time (TT), can differ-entiate between exogenous and endogenous pathway effects anddistinguish them from effects on fibrin formation. Typically, theinfluence of compounds on plasmatic blood coagulation is deter-mined by measuring PT, aPTT, and TT ex vivo.

2.1.2. Procedure Male Sprague-Dawley rats weighing 200–220 g are administeredtest compound, or vehicle as a control, through an oral, intraperi-toneal, or intravenous route. After a period of time for absorption(adsorption time), animals are anesthetized by intravenous injec-tion of sodium pentobarbital (60 mg/kg). The caudal caval vein isexposed by midline incision and 1.8 ml of blood is collected into

In Vitro Methods of Evaluating Antithrombotics and Thrombolytics 3

a plastic syringe containing 0.2 ml of 100 mM citrate buffer, pH4.5 (Behring Werke, Marburg, DE). The sample is immediatelyagitated and then subjected to centrifugation in a plastic tube at1,500×g for 10 min, after which plasma is collected and trans-ferred to a clean plastic tube. Coagulation tests for TT, PT, andaPTT should be performed within 3 hours (h).

In general, citrated plasma coagulates upon the addition ofthe appropriate compound (see below), and time to clot for-mation (coagulation time) is determined using a coagulometer(Schnittger and Gross, Amelung, Brake, DE). For the detailedlaboratory diagnosis of bleeding disorders and assessment ofblood coagulation, see Palmer (4) and Nilsson (5).

PT. PT measures effects on the exogenous pathway of coag-ulation. Citrated plasma (0.1 ml) is incubated for 1 min at 37◦C,at which point 0.2 ml of human thromboplastin (Thromborel R©;Behringwerke) is added. The coagulometer is started, and time toclot formation is recorded.

aPTT. aPTT measures effects on the endogenous pathway ofcoagulation. Citrated plasma (0.1 ml) is mixed with 0.1 ml ofhuman placenta lipid extract (Pathrombin R©; Behringwerke), andthe mixture is incubated for 2 min at 37◦C. Coagulation is ini-tiated by the addition of 0.1 ml of 25 mM calcium chloride, atwhich point the coagulometer is started and time to clot forma-tion is recorded.

TT. TT measures effects on fibrin formation. Citrated plasma(0.1 ml) is mixed with 0.1 ml of diethyl barbiturate–citrate buffer,pH 7.6 (Behringwerke), and the mixture is incubated for 1 min at37◦C. Bovine test thrombin (0.1 ml) (30 IU/ml; Behringwerke)is added, at which point the coagulometer is started, and the timeto clot formation is recorded.

2.2.Thrombelastography


Thrombelastography (TEG) was developed by Hartert in 1948(6). The thrombelastograph is a device that provides a contin-uous recording of the process of blood coagulation and subse-quent clot retraction. Blood samples are transferred to cuvettesand maintained at 37◦C. The cuvettes are set in motion aroundtheir vertical axes. Initially, a mirror suspended by a torsion wire inthe plasma remains immobile as long as the plasma is fluid. Thereis a dynamic interplay between the cuvette and the mirror as fibrinforms, resulting in the transmission of motion within the cuvetteto the mirror. The mirror will oscillate, the amplitude of whichis governed by the specific mechanical properties of the clot, andreflect light onto a thermo-paper recording. Modern thrombelas-tographs translate the light recording into a digital signal that canbe readily analyzed using a computer program.

4 Mousa

2.2.2. Procedure TEG can be performed with whole blood or citrated platelet-rich or platelet-poor plasma after re-calcification. Blood samplesare obtained from test animals (i.e. Beagle dogs, 12–20 kg; rab-bits, 1.7–2.5 kg; Wistar rats, 150–300 g; or humans). Test sub-jects receive the compound of interest by intravenous (iv), sub-cutaneous (sc) or oral administration. Ten- or twenty-min postdosing (iv or sc administration), or 60-, 90- or 180-min postdosing (oral administration), blood is collected. The blood sam-ples are mixed with 3.8% trisodium citrate solution (1:9 citratesolution:blood) as an anticoagulant. Citrated whole blood is re-calcified by the addition of 0.4 ml of isotonic calcium chloride.An aliquot (0.36 ml) of re-calcified whole blood is transferred toa pre-warmed cup of the thrombelastograph. After the apparatushas been correctly adjusted and the samples have been sealed withliquid paraffin to prevent drying, the start time for the procedureis noted, and the thrombelastogram is recorded for 2 h.

2.2.3. Evaluation The following measurements are the standard variables of TEG:Reaction time (r). The time from sample placement in the

cup until onset of clotting (defined as an amplitude of 1 mm).This represents the rate of initial fibrin formation.

Clot formation time (k). The difference from the 1 mm r to20 mm amplitude. k represents the time for a fixed degree ofviscoelasticity to be achieved by the clot formation due to fibrinbuildup and cross linking.

Alpha angle (α◦). Angle formed by the slope of the TEG trac-ing from the r to k value. It denotes the speed at which solid clotforms.

Maximum amplitude (MA). Greatest amplitude on the TEGtrace. MA represents the absolute strength of the fibrin clot and isa direct function of the maximum dynamic strength of fibrin andplatelets.

Clot strength (G; in dynes per square centimeter). It is definedby G = (5,000MA)/(96–MA). In tissue factor-modified TEG(7), clot strength is clearly a function of platelet concentration.

Lysis 30, Lysis 60 (Ly30, Ly60). Reduction of amplitude rela-tive to MA at 30 and 60 min after the time of MA. These param-eters represent the influence of clot retraction and fibrinolysis.

Readers are referred to a number of studies in which TEGhas been instrumental in advancing the field of antithromboticsand thrombolytics. Most recently, TEG has been used to analyzethe effects of a variety of stimuli on platelet/fibrin clot dynam-ics (8). In other work, Bhargava et al. (9) used TEG to com-pare the anticoagulant effects of a new potent heparin preparationand then commercially available heparin in vitro using citrateddog and human blood. Barabas et al. (10) used the fibrin plate


assay and TEG to assess the anti-fibrinolytic effects of syntheticthrombin inhibitors. Scherer et al. (11) described a short-time,endotoxin-induced rabbit model of hyper-coagulability for thestudy of the coagulation cascade, enabling the analysis of coag-ulation inhibitors and their therapeutic effects by a number oftechniques, including TEG.

In 1997, Khurana et al. (7) introduced tissue factor-modifiedTEG to study platelet GPIIb/IIIa function, establishing a quan-titative assay of platelet function. Using this modification to TEG,Mousa et al. (8) identified two classes of GPIIb/IIIa antagonists,one with high binding affinity for resting and activated plateletsand a slow platelet dissociation rate (class I) that exhibited potentinhibition of platelet function, and one with a fast platelet dissoci-ation rate (class II). TEG was also used in a phase-II clinical trialto assess the efficacy of an oral platelet GPIIb/IIIa antagonist onplatelet/fibrin clot dynamics (12).

Zuckerman et al. (13) compared TEG with other commoncoagulation tests (fibrinogen, PT, aPTT, platelet count, and fibrinsplit products) and found a strong correlation between thrombe-lastographic variables and common laboratory tests. Moreover,TEG had a higher sensitivity for blood clotting anomalies. TEGalso provides additional information on the hemostatic process.In contrast to most laboratory assays, in which the end point isthe formation of the first fibrin strands, TEG measures the coag-ulation process from the initiation of clotting to the final stagesof clot lysis and retraction. Another advantage of TEG is that itallows the use of whole non-anticoagulated blood without theinfluence of citrate or other anticoagulants.

2.3. Chandler Loop


The Chandler loop technique measures the generation of in vitrothrombi in a moving column of blood (14). Thrombi gener-ated in the Chandler device are morphologically similar to humanthrombi formed in vivo (15), with platelet-rich upstream sections(“white heads”) that are relatively resistant to tissue plasminogenactivator(t-PA)-mediated thrombolysis as compared to red bloodcell-rich downstream components (“red tails”) (16).

2.3.2. Procedure One millimeter of non-anticoagulated whole blood is drawndirectly into a polyvinyl tube 25 cm in length with an internaldiameter of 0.375 cm (1 mm=9.9-cm tubing). The two endsof the tube are then brought together and closed using an out-side plastic collar. The circular tube is placed at the center of aturntable, tilted to an angle of 23º, and then rotated at 17 rpm.When the developing thrombus inside the tube becomes largeenough to occlude the lumen, the blood column becomes staticand begins to move on the table in the direction of rotation.

6 Mousa

Stringer et al. (16) used this method to determine the influ-ence of an anti-plasminogen activator inhibitor (PAI)-1 antibody(CLB-2C8) on t-PA-induced lysis of Chandler thrombi in vitro.Investigators used citrated blood supplemented with 5.8 μM[125I]-labeled fibrinogen prior to re-calcification. Thrombi gen-erated in the Chandler loop were washed with isotonic salineand then cut transversally into an upstream (head) and a down-stream (tail) part, and then each part was analyzed using a gammacounter to determine pre-values. The head and tail were then sub-jected to thrombolysis by the addition of 300 μl of phosphate-buffered saline (PBS) containing plasminogen (2 μM) and t-PA(0.9 nM). Over a period of 240 min, aliquots (10 μl) from eachsection were removed at 30, 60, 120, 180, and 240 min, andradioactivity was determined. Radioactivity at each time point ascompared to the pre-value was then expressed as a percentage ofclot lysis.

In 1998, Van Giezen et al. (17) used this method to success-fully differentiate the effects of an anti-PAI-1 polyclonal antibody(PRAP-1) on human and rat thrombi.

2.4. PlateletAggregation andDe-aggregation inPlatelet-Rich Plasmaor Washed Platelets(Born Method)


Contact between non-activated platelets and exposed sub-endothelial tissue leads to adhesion through two main mecha-nisms: (1) at high shear rates, binding of sub-endothelial vWFto platelet GPIb–IX–V complex and (2) binding of collagen tointegrin α2β1 and GPVI. Platelet adhesion initiates several pro-cesses, including shape changes, secretion, and the activation ofGPIIb/IIIa ligand binding sites, resulting in the formation ofplatelet aggregates. Activation of GPIIb/IIIa is also achievedthrough receptor cross-signaling initiated by the binding of anumber of agonists to G-protein-coupled receptors. To mea-sure platelet aggregation, one of the following agonists is addedto platelet-rich plasma (PRP) or washed platelets (WP): ADP,arachidonic acid (which is converted to thromboxane A2) or thethromboxane agonist U46619, collagen, thrombin or thrombinreceptor-activating peptide (TRAP), serotonin, epinephrine, orplatelet activating factor (PAF). Upon stirring, the formation ofplatelet aggregates is monitored photometrically as changes inoptical density, typically for 4 min. This test was developed byBorn (18, 19) and is used to quantitatively evaluate the effect ofcompounds on the induction of platelet aggregation in vitro orex vivo. For in vitro studies, human PRP is the preferred startingmaterial.


2.4.2. Procedure For ex vivo assays, mice, rats, or guinea pigs of either sex receivetest compound, or vehicle as a control, by oral, intraperitoneal(ip), or iv administration. At the end of absorption time, bloodis collected by caval venipuncture under pentobarbital sodiumanesthesia with xylazine (8 mg/kg intramuscular) premedication.From rabbits (Chinchilla, 3 kg in weight), blood is drawn by car-diopuncture under xylazine (20 mg/kg intramuscular) sedation.A control blood sample is collected before administration of thetest compound, and the second sample is drawn at the end ofabsorption time. For in vitro assays of human blood, samples arecollected from the vein of adult volunteers who have not receivedany medication during the 2 weeks prior to collection.

2.4.2.1. Preparationof PRP, Platelet-poorPlasma (PPP), and WP

The entire procedure is performed in plastic (polystyrene) tubesand carried out at room temperature. Freshly collected venousblood is anticoagulated with hirudin (1:9, hirudin:blood) or anti-coagulant citrate dextrose (ACD) solution (1:9, ACD:blood) andthen subjected to centrifugation at 170×g for 15 min to obtainPRP. The PRP supernatant is carefully removed, and the sample issubjected to centrifugation again at 1,500×g for 10 min to obtainPPP. PRP is diluted with PPP to a platelet count of 3×108/mlbefore use in the aggregation assay. To obtain WP, 8.5 volumesof human blood are collected into 1.5 volumes of ACD and thensubjected to centrifugation, as described for PRP. PRP is acidi-fied to a pH of 6.5 by the addition of ACD (approximately 1 mlper 10 ml of PRP). Acidified PRP is subjected to centrifugationfor 20 min at 430×g, and then the pellet is re-suspended to theoriginal volume in Tyrode’s solution (120 mM NaCl, 2.6 mMKCl, 12 mM NaHCO3, 0.39 mM NaH2PO4, 10 mM HEPES,5.5 mM Glucose, and 0.35% albumin) and diluted to a plateletcount of 3×108/ml. The assay should be completed within 3 hof blood collection.

For ex vivo assays, duplicate samples of PRP (320 μl) fromdrug-treated and vehicle control subjects (for rabbits, the con-trol sample is taken before drug administration) are inserted intothe aggregometer at 37◦C under continuous magnetic stirring at1,000 rpm. After the addition of 40 μl of physiological saline and40 μl of aggregating agent, changes in optical density are moni-tored continuously at 697 nm.

For in vitro assays, 40 μl of the test solution is added to320 μl of PRP or WP from untreated subjects. The samples areinserted into the aggregometer and incubated at 37◦C for 2 minunder continuous magnetic stirring at 1,000 rpm. After the addi-tion of 40 μl of aggregating agent, changes in optical density aremonitored continuously at 697 nm for 4 min, or until aggre-gation values are constant. In cases of thrombin activation ofPRP, glycine–proline–aspartate–proline (GPRP) peptide is added

8 Mousa

in order to prevent fibrin formation. To measure de-aggregation,experimental compounds are added to stimulated PRP at 70%or 100% of control aggregation, and samples are monitored for10 min. De-aggregation is measured as the decrease in light trans-mission over time (20).

2.4.3. Evaluation

2.4.3.1. For In VitroAssays

1. Percent inhibition of platelet aggregation is determined ineach concentration group relative to the respective vehiclecontrol.

2. IC50 values are determined from non-linear fitting of theconcentration–effect relationship curve. IC50 is defined asthe concentration of test drug that achieves half-maximalinhibition of aggregation.

3. Percent de-aggregation is determined 10 min after theaddition of compound; IC50 is calculated from theconcentration–effect curve

4. Statistical significance is evaluated by means of the unpairedStudent’s t-test.

2.4.3.2. For Ex VivoAssays

1. The mean values for aggregation in each dosage group arecompared to the respective vehicle control group (for rab-bits, the control is before drug administration).

2. ED50 values are determined from the dose–response curves.ED50 is defined as the dose of drug that achieves half-maximal inhibition of aggregation in animals.

3. Statistical significance is evaluated by means of the Student’st-test (paired for rabbits; unpaired for others).

2.4.4. CriticalAssessment of theMethod

This assay, introduced by Born in 1926, has become a standardmethod in the clinical diagnosis of platelet function disorders andaspirin intake. Furthermore, the method is widely used in thediscovery of antiplatelet drugs. The advantages of the methodinclude the ability to rapidly measure a functional parameter inintact human platelets. However, processing of platelets duringthe preparation of PRP, WP, or filtered platelets from whole bloodcan result in platelet activation and separation of large platelets.

2.4.5. Modifications ofthe Method

Several authors have described modifications of the Born assay.Breddin et al. (21) described the use of a rotating cuvette to mea-sure spontaneous aggregation of platelets from vascular patients.Klose et al. (22) measured platelet aggregation under laminarflow conditions using a thermo-regulated cone-plate streamingchamber in which shear rates were continuously augmented andplatelet aggregation was measured based on light transmissionthrough a transillumination system. Marguerie et al. (23, 24)


developed a method of measuring two phases of platelet aggre-gation after gel filtration of a platelet suspension (see below).Lumley and Humphrey (25) described a method to measureplatelet aggregation in whole blood (see below). Fratantoni andPoindexter (26) used a microtiter plate reader with specific mod-ifications of sample agitation to measure platelet aggregation.A comparison of the 96-well microtiter plate method and con-ventional aggregometry revealed similar dose–response curves forthrombin, ADP, and arachidonic acid. Ammit and O’Neil (27)used a quantitative bioassay of platelet aggregation for rapid andselective measurement of PAF.

2.5. PlateletAggregation After GelFiltration


Triggering of platelet activation by low concentrations of ADP,epinephrine, or serotonin, the so-called weak platelet agonists, inplasma- and fibrinogen-free platelet suspensions does not resultin platelet aggregation unless exogenous fibrinogen is added. Incontrast, platelet aggregation induced by thrombin, collagen, orprostaglandin endoperoxide, so-called strong agonists, is inde-pendent of exogenous fibrinogen because these substances inducethe secretion of intracellular platelet ADP and fibrinogen. Anal-ysis of platelet aggregation in gel-filtered platelet samples is car-ried out in cases when fibrinogen or vWF is needed in a definedconcentration, or when plasma proteins could negatively interferewith the effects of compounds. The assay is used primarily to eval-uate the influence of compounds on platelet integrin GPIIb/IIIaor other integrins or on platelet GPIb–IX–V.

2.5.2. Procedure

2.5.2.1. Preparationof Gel-Filtered Platelets

The entire procedure is performed in plastic (polystyrene) tubes atroom temperature (24). Blood is drawn from healthy adult vol-unteers who have received no medication in the 2 weeks priorto collection. Venous blood (8.4 ml) is collected into 1.4 mlof ACD solution and subjected to centrifugation for 10 min at120×g. PRP is carefully removed, the pH is adjusted to 6.5 withACD solution, and the sample is subjected to centrifugation againat 285×g for 20 min. The resulting pellet is re-suspended inTyrode’s buffer (approx. 500 μl of buffer/10 ml of PRP), and theplatelet suspension is applied immediately to a Sepharose CL 2Bcolumn. Equilibration and elution (flow rate, 2 ml/min) are donewith Tyrode’s buffer without hirudin and apyrase. Platelets arerecovered in the void volume. The platelet suspension is adjustedto a final cell concentration of 4×108/ml. Gel-filtered platelets(GFP) are kept at room temperature for 1 h before the assay isstarted.

10 Mousa

2.5.2.2. ExperimentalCourse

For aggregation studies, GFP in Tyrode’s buffer are incubatedwith CaCl2 (final concentration, 0.5 mM) with or without fib-rinogen (final concentration, 1 mg/ml) in polystyrene tubes.After 1 min, 20 μl of the test compound, or vehicle as a con-trol, are added and the sample is incubated for an additional2 min. Platelet agonist is added (20 μl), and changes in lighttransmission are recorded. The entire procedure is done undercontinuous magnetic stirring at 37◦C (1,000 rpm) in the aggre-gometer. Samples with CaCl2 but without fibrinogen confirm thatplasma proteins have been properly filtered out if neither sponta-neous aggregation nor aggregation in the presence of weak ago-nists occurs; full aggregation of GFP in response to 10 μM ADPconfirms that platelets are intact (i.e., minor pre-activation by gelfiltration).

Readers are referred to several references for a detailedmethodology and evaluation of different agents by gel filtrationplatelet aggregation (23, 24, 28).

2.6. PlateletAggregation in WholeBlood


This method uses a whole blood platelet counter, which countssingle platelets and does not require separation of platelets fromother blood cell types. Platelet aggregation is induced in anticoag-ulated human whole blood samples by addition of the aggregat-ing agents arachidonic acid or collagen. The number of plateletsis determined in drug-treated and vehicle control samples andthe percent inhibition of aggregation and IC50 values are cal-culated for each dosage group. This test system enables assess-ment of the effect of compounds on other blood cells, which canindirectly influence platelet aggregation. The method was origi-nally described by Lumley and Humphrey (25) and Cardinal andFlower (29).

2.6.2. Procedure The entire procedure is performed in plastic (polystyrene) tubesand is carried out at room temperature. Blood is drawn fromhealthy adult volunteers who have not received medication duringthe 2 weeks prior to collection. Venous blood (9 ml) is anticoagu-lated with 1 ml of sodium citrate and maintained in a closed tubeat room temperature for 30–60 min until the start of the test.

For aggregation analysis, 10 μl of compound, or vehicle asa control, are added to 480 μl of citrated blood. Samples inclosed tubes are pre-incubated for 5 min in a 37◦C water shakerbath with shaking (75 strokes/min). Aggregating agent (10 μl) isadded and samples are incubated for another 10 min. The num-ber of platelets (platelet count) is determined in aliquots of 10 μlimmediately before and 10 min after the addition of aggregating


agent (initial platelet count and 10-min platelet count, respec-tively) using a hemocytometer.

The following samples are prepared in duplicate:1. Control aggregation/spontaneous aggregation: 480 ml

blood + 20 ml vehicle. Samples with >20% spontaneousaggregation should not be used to test for inducedaggregation.

2. Maximum aggregation: 480 ml blood + 10 ml vehicle+ 10 ml aggregating agent. Represents the maximuminduced aggregation rate.

3. Test substance aggregation: 480 ml blood + 10 ml test sub-stance + 10 ml aggregating agent.

2.7. Platelet Micro-andMacro-AggregationUsing Laser


A new platelet aggregometer (AG-10; Kowa, Japan) that usesa laser light-scattering beam was introduced in 1996 by Tohgiet al. (30). This technique is a highly sensitive method of studyingplatelet aggregation based on the measurement of mean radius orparticle size, making it possible to record the kinetics of formationof micro- and macro-aggregates in real time. Sensitivity in mea-surement of spontaneous aggregation is higher than in routinelight transmittance.

Platelet aggregate size, derived from the total voltage of light-scatter intensity at 1.0-second (s) intervals over a 10-min period,can be divided into three ranges: small (diameter 9–25 μm),medium (26–50 μm), and large (>50 μm) aggregates. Usinglaser scatter aggregation, it was found that young smokers hadan increased number of small platelet aggregates, which were un-detectable using a conventional aggregometer and the turbido-metric method (31). The light-scatter aggregometer can detectplatelet aggregation in the small-aggregate size range after theaddition of unfractionated heparin (UFH), and the aggregates aredissociated upon incubation with protamine sulfate. Analysis ofplatelet aggregation induced by 0.5 U/ml of UFH in 36 normalsubjects with no history of heparin exposure revealed 13 subjectswith a positive response in excess of 0.5 V of light intensity in thesmall-aggregate size range. In chronic hemodialysis patients whohad used heparin regularly for many years, a positive response,that is, the detection of heparin-induced aggregates, was observedin 37 of 59 patients, an increase over normal subjects. Light inten-sity in the small-aggregate size range was also enhanced duringheparinized dialysis. In patients with a positive heparin response,the intensity of aggregate formation after heparin was significantlyhigher than heparin non-responders. Using the same system, it

12 Mousa

has been shown that enhanced platelet aggregation response toheparin is not inhibited by aspirin or argatroban, but can be inhib-ited by anti-GPIIb/IIIa antibodies. Enhanced platelet aggrega-tion during heparin infusion was observed without the additionof ADP or TRAP using laser aggregometry (32).

Limitations. This technique cannot be applied to wholeblood, but can be used with PRP, WP, or GFP.

2.8. FibrinogenReceptor Binding


This assay is used to evaluate the binding characteristics of drugsthat target the fibrinogen receptor. A single concentration of radi-oligand (125I-fibrinogen, 30–50 nM) is incubated with humanGFP in the presence of increasing concentrations of a non-labeledtest drug (0.1 nM–1 mM). If the test drug binds to the fibrinogenreceptor, it will compete with the radiolabeled ligand for receptorbinding sites. Generally, the higher the affinity of the test drug forthe receptor, the lower the concentration required to compete forbinding, and the more potent the test drug will be. Platelets areactivated by 10 mM ADP to stimulate 125I-fibrinogen binding tothe GPIIb/IIIa receptor.

2.8.2. Procedure

2.8.2.1. Preparationof GFP

Blood is collected from a healthy volunteer (200 ml). An aliquot(8.4 ml) is mixed with 1.4 ml of ACD buffer in a polystyrol tubeand then subjected to centrifugation at 1,000 rpm for 15 min.The resulting PRP is collected and an aliquot is removed for aplatelet count. Ten ml of PRP are mixed with 1 ml of ACD bufferto yield ACD–PRP, pH ∼6.5, and then 5 ml portions of ACD–PRP are transferred to plastic tubes and subjected to centrifu-gation at 1,600 rpm for 20 min. The resulting supernatants aredecanted, and each pellet is re-suspended in 500 μl of Tyrodebuffer C. An aliquot is removed for a platelet count to determineplatelet loss, and then the platelet suspension is transferred to aSepharose column that has been pre-equilibrated with approxi-mately 100 ml of degassed Tyrode buffer B (flow rate, 2 ml/min).The column is closed, and sample is eluted with degassed Tyrodebuffer B (flow rate, 2 ml/min). The first platelets appear in the18–20 min fractions and are collected thereafter for 10 min in aclosed plastic cup. GFP are reconstituted at a density of 4×108

platelets/ml with Tyrode buffer B and maintained at room tem-perature until use.


Each concentration of drug is tested in triplicate using No. 72708Sarstedt tubes. The total volume of each test sample is 500 μl.The concentration of 125I-fibrinogen is constant for all samples(10 μg/500 μl).


2.8.2.3. CompetitionExperiments

Competition reactions are carried out using a negative control(i.e., double distilled water), non-labeled fibrinogen, and increas-ing concentrations of test compound, as follows:

• 100 μl 125I-fibrinogen• 100 μl non-labeled fibrinogen or test drug (increasing con-

centrations of 10−10 to 10−3 M)• 50 μl ADP

2.8.2.4. Non-SpecificBinding

Non-specific binding of 125I-fibrinogen is defined as the level ofbinding of radioligand in the presence of 10−5 M non-labeledfibrinogen. The binding reaction is initiated by the addition of250 μl of GFP (4×108 platelets/ml). The samples are incubatedfor 30 min at room temperature, and then 100 μl is transferred toa microcentrifuge tube containing 400 μl of a glucose solution.The tubes are centrifuged at 11,750 rpm for 2 min to separatebound 125I-fibrinogen from free radioligand. The supernatant iscarefully decanted and allowed to evaporate for approximately30 min. The radioactivity of the platelet pellet is measured for1 min using a gamma counter (efficiency in the range of 65.3%).

Readers are referred to several excellent references for adetailed methodology and evaluation of various mechanismsand agents assessed using the fibrinogen binding assay (23, 24,33, 34).

2.9. Euglobulin ClotLysis Time


Euglobulin lysis time is used as an indicator of the influence ofcompounds on fibrinolytic activity in rat blood and is based onthe procedure of Gallimore et al. (35). The euglobulin fractionof plasma is separated from inhibitors of fibrinolysis by acid pre-cipitation and centrifugation. Euglobulin predominantly consistsof plasmin, plasminogen, plasminogen activator, and fibrinogen.Addition of thrombin to this fraction results in the formation offibrin clots. The lysis time of these clots is related to the activityof activators of fibrinolysis (e.g., plasminogen activators). Thus,compounds that stimulate the release of t-PA from the vessel wallcan be detected with this assay.

2.9.2. Procedure Rats are anesthetized by ip injection of pentobarbital sodium(60 mg/kg) and placed on a heating pad (37◦C). At the sametime, the test solution, or vehicle as a control, is administered ivor ip. Twenty-five minutes later, the animals receive another ipinjection of sodium pentobarbital (12 mg/kg) to maintain themin deep narcosis for 45 min.

2.9.2.1. PlasmaPreparation

After the test compound is absorbed, the inferior caval vein isexposed by a midline excision and blood (1.8 ml) is drawn usinga plastic syringe containing 0.2 ml of a 3.8% sodium citrate

14 Mousa

solution. The sample is thoroughly mixed, transferred to a plas-tic tube, and immediately immersed in ice. Plasma is obtained bycentrifugation at 2,000×g for 10 min at 2◦C.

2.9.2.2. EuglobulinPreparation

A 0.5 ml aliquot of plasma is added to 9.5 ml of ice-cold dis-tilled water, and then the pH is brought to 5.3 by the addition of0.13 ml of 1% acetic acid. The diluted plasma is kept on ice for10 min and then the precipitated euglobulin fraction is collectedby centrifugation at 2,000×g for 10 min at 2◦C. The supernatantis discarded and the remaining fluid is removed by wicking awayexcess moisture with filter paper for 1 min. The euglobulin pre-cipitate is dissolved in 1 ml of 0.12 M sodium acetate.

2.9.2.3. Euglobulin LysisAssay

Aliquots (0.45 ml) of the euglobulin solution are transferred totest tubes, and 0.05 ml of thrombin (Test Thrombin, Behring-werke) (25 U/ml) is added. The tubes are transferred to a waterbath at 37◦C. The time interval between the addition of thrombinand complete lysis of the clots is recorded.

2.10. Flow Behaviorof Erythrocytes


The deformation of erythrocytes is an important rheological phe-nomenon in blood circulation (36). It allows the passage of nor-mal red cells through capillaries with diameters smaller than thatof the discoid cells and reduces the bulk viscosity of blood flow-ing in large vessels. In this test, the initial flow of filtrate is used asa measure of erythrocyte deformability. Prolonged filtration timecan be attributed to two fundamental pathological phenomena:an increased rigidity of individual red cells and an increased ten-dency of the cells to aggregate. To simulate decreased red bloodcell deformability, the erythrocytes are artificially modified by one(or a combination) of the following stress factors:

• addition of calcium ions (increase in erythrocyte rigidity)• addition of lactic acid (decrease in pH value)• addition of 350–400 mmol NaCl (hyper-osmolarity)• storing the sample for at least 4 h (cellular ageing, depletion

of ADP)This test is typically used to evaluate the effect of test compoundson the flow behavior of erythrocytes.

2.10.2. Procedure Apparatus. Erythrocyte filtrometer model MF 4 (Fa. Myrenne,52159 Roetgen, Germany) equipped with a membrane filter(Nuclepore Corp., Pleasanton, CA, USA) (pore diameter, 5–10 μm; pore density, 4×105 pores/cm2).

Ex vivo. Blood is collected from Beagle dogs (12–20 kg), rab-bits (1.2–2.5 kg) or Wistar rats (150–300 g). Animals receive thetest compound by oral, sc or iv administration 15, 60, 90, or180 min before the withdrawal of blood.


In vitro. Following the addition of test compound, bloodis incubated at 37◦C for 5 or 30 min. Freshly collected venousblood is anticoagulated with K-EDTA (1 mg/ml) or heparin(5 IU/ml heparin sodium) and then subjected to centrifugationat 3,000 rpm for 7 min. The supernatant (plasma) and buffycoat are removed and discarded. The packed erythrocytes are re-suspended in autologous plasma containing 0.25% human albu-min, and the haematocrit is adjusted to 10%. Red blood cells arethen subjected to one of a combination of stress factors men-tioned above.

A portion (2 ml) of the stressed suspension is applied to thefiltrometer and initial flow rate is determined. The filtration curveis plotted automatically.

2.11. Filterabilityof Erythrocytes


The single erythrocyte rigidometer (SER) measures the deforma-bility of individual red blood cells by measuring passage timethrough a pore under constant shear stress. In this assay, thepassage times of single erythrocytes through one pore in a syn-thetic membrane are determined (37–39). The pore in the mem-brane essentially represents a capillary with a defined diameter andlength. The driving pressure is produced by constant shear stress.The passage of red blood cells is measured based on changesin electrical current. For example, a constant current of 50–200nA is applied, and passage of an erythrocyte through the pore isrecorded as an interruption in current. This test is used to detectcompounds that improve the filterability, and thus deformability,of erythrocytes. To simulate decreased red blood cell deformabil-ity, the erythrocytes are artificially modified by one or a combina-tion of the following stress factors:

• calcium ions (increase erythrocyte rigidity)• lactic acid (decreases pH value)• 350–400 mmol NaCl (generates a state of hyper-osmolarity)• storage for at least 4 h (cellular ageing and depletion of ADP)

2.11.2. Procedure Apparatus. Single erythrocyte rigidometer (Myrenne, 52159Roetgen, Germany).

Ex vivo. Blood is collected from Beagle dogs (12–20 kg), rab-bits (1.2–2.5 kg), or Wistar rats (150–300 g), or from human.The subject receives the test compound by oral, sc, or iv adminis-tration 15, 60, 90, or 180 min before the withdrawal of blood.

In vitro. Following the addition of the test compound, bloodsamples are incubated at 37ºC for 5 or 30 min.

Blood samples are mixed with K-EDTA (1 mg/ml blood) orheparin (5 IE/ml heparin sodium) to prevent clotting and thensubjected to centrifugation at 3,000 rpm for 7 min. The plasma

16 Mousa

and buffy coat are removed and discarded. The packed erythro-cytes are re-suspended in filtered HEPES buffer containing 0.25%human albumin and the haematocrit value is adjusted to <1%.The red blood cells are altered by one or a combination of stressfactors, and then a portion (2 ml) of the stressed suspension isapplied to the rigidometer. The passage time of a population of250 erythrocytes (tm) is recorded. Cells that remain in the porefor more than 100 ms (tm > 100 ms) cause rheological occlusion.Untreated red blood cells serve as the control for this assay.

2.12. ErythrocyteAggregation


The aggregation of red blood cells into rouleaux and the transi-tion of rouleaux into three-dimensional cell networks is a rheo-logical parameter that decisively influences the flow behavior ofblood, especially in disturbed microcirculation. In this test, anerythrocyte aggregometer is used to measure erythrocyte aggre-gation. The transparent measuring chamber in a cone-and-plateconfiguration is transilluminated by light of a defined wave length,and the intensity of the transmitted light, which is modified by theaggregation process, is recorded. The method has been used suc-cessfully to determine the effect of test compounds on erythrocyteaggregation (37, 40).

Apparatus. SER (Fa. Myrenne, 52159 Roetgen, Germany).Ex vivo. Blood is collected from Beagle dogs (12–20 kg), rab-

bits (1.2–2.5 kg), or Wistar rats (150–300 g). The animals receivetest compound by oral, sc, or iv administration 15, 60, 90, or180 min before the withdrawal of blood.

In vitro. Following the addition of test compound, the bloodsample is incubated at 37◦C for 5 or 30 min. Blood is obtainedby venipuncture and mixed with K-EDTA (1 mg/ml) or hep-arin (5 IU/ml heparin sodium) to prevent clotting. Erythrocyteaggregation is determined in whole blood with a haematocrit of40%. A portion (40 μl) of the blood is transferred to the measur-ing device and the red cells are dispersed at a shear rate of 600/s.After 20 s, flow is switched to stasis, and the extent of erythrocyteaggregation is determined photometrically.

3. In Vitro Modelsof Thrombosis

A variety of methods have been used to assess the ex vivo and/orin vitro efficacy of platelet antagonists, including photometricaggregometry, whole blood electrical aggregometry, and parti-cle counting, as described earlier. In photometric aggregometry,a sample is placed in a stirred cuvette in the optical light path


between a light source and a light detector. Aggregate forma-tion is monitored by a decrease in turbidity, and the extent ofaggregation is measured as percent maximal light transmission.The major disadvantage of this technique is that it cannot beapplied to whole blood, since the presence of erythrocytes inter-feres with optical detection. Furthermore, it is insensitive to theformation of small aggregates. Particle counters are used to quan-titate the size and the number of particles suspended in an elec-trolyte solution by monitoring the electrical current between twoelectrodes immersed in the solution. Aggregation in this system isquantitated by counting the number of platelets before and afterstimulation and is usually expressed as a percentage of the ini-tial count (41). The disadvantage of this technique is that it can-not distinguish platelets and platelet aggregates from other bloodcells of the same size. Thus, one is limited to counting only afraction of single platelets as well as aggregates that are muchlarger than erythrocytes and leukocytes. Electrical aggregometryallows the detection of platelet aggregates as they attach to elec-trodes immersed in a stirred cuvette containing whole blood or aplatelet suspension. Attachment results in a decrease in conduc-tance between the two electrodes, which can be quantitated inunits of electrical resistance. A disadvantage of this method is thatit is not sensitive enough to detect small aggregates (42).

This section will discuss two complementary in vitro flowmodels of thrombosis that can be used to accurately quantifyplatelet aggregation in anticoagulated whole blood and eval-uate the inhibitory effect of platelet antagonists: (1) a visco-metric flow cytometric assay that measures direct shear-inducedplatelet aggregation in the bulk phase (43) and (2) a parallel-plateperfusion chamber assay coupled with a computerized videomi-croscopy system that quantifies the adhesion and subsequentaggregation of human platelets in anticoagulated whole bloodflowing over an immobilized substrate (i.e., collagen I) (43, 44).We also discuss a third in vitro flow assay described by Mousa et al.(44) in which surface-anchored platelets are pre-incubated with aGPIIb/IIIa antagonist, unbound drug is washed away, and thenTHP-1 monocytic cells are perfused into the system, enablingthe characterization of agents with markedly distinct affinities andreceptor-bound lifetimes.

3.1. Cone-and-PlateViscometry UnderShear-FlowCytometry


The cone-and-plate viscometer is an in vitro flow model used toinvestigate the effects of bulk fluid shear stress on suspended cells.Anticoagulated whole blood (or isolated cell suspension) is placed

18 Mousa

between the two stainless steel plates of the viscometer. Rotationof the upper conical plate causes a well-defined and uniform shearstress to be applied to the entire fluid medium (1). The shear rate(γ ) in this system can be readily calculated from the cone angleand the speed of the cone using the following formula:

γ =(

2πω

60θcp

),

where γ is the shear rate per second, ω is the cone rotational ratein revolutions per minute (rev/min) and θcp is the cone angle inradians. The cone angle is typically in the range of 0.3–1.0◦. Shearstress, τ , is proportional to shear rate, γ , as shown by

τ = μγ ,

where μ is the viscosity of the cell suspension (the viscosity ofanticoagulated whole blood is approximately 0.04 cp at 37◦C).Rotational viscometers are capable of generating shear stress inthe range of ∼2 dyn/cm2 (venous level) to greater than 200dyn/cm2 (stenotic arteries).

3.1.2. Procedure Single platelets and platelet aggregates generated in blood uponshear exposure are differentiated from other blood cells by theircharacteristic forward-scatter and fluorescence profiles from flowcytometric analysis using fluorophore-conjugated platelet-specificantibodies (43). This technique requires no washing or centrifu-gation steps that could potentially induce artifacts due to plateletactivation and enables the analysis of platelet function in the pres-ence of other blood elements.

3.1.2.1. Isolation ofHuman Platelets

Venous blood is drawn by venipuncture into polypropylenesyringes containing either sodium citrate (0.38% final concentra-tion) or heparin (10 U/ml final concentration). Anticoagulatedwhole blood is subjected to centrifugation at 160×g for 15 minto obtain PRP.

3.1.2.2. Isolation of WP PRP is subjected to centrifugation again at 1,100×g for 15 min inthe presence of 2 μM PGE1. The platelet pellet is re-suspended inHEPES-Tyrode buffer containing 5 mM EGTA and 2 μM PGE1.Platelets are washed and collected by centrifugation (1,100×g for10 min) and then re-suspended in HEPES-Tyrode buffer at a celldensity of 2×108 cells/ml and maintained at room temperaturefor no longer than 4 h before use in aggregation/adhesion assays.


The steps described in this section outline the procedure used toquantify platelet aggregation induced by shear stress in the bulkphase as well as the inhibitory effects of platelet antagonists. Fora detailed description, see Konstantopolous et al. (43).


1. Incubate anticoagulated whole blood with platelet antago-nist, or vehicle as a control, at 37◦C for 10 min.

2. Place the sample (typically 500 μl) on the stationary plate ofa cone-and-plate viscometer maintained at 37◦C.

3. Remove a small aliquot (∼3 μl) from the pre-sheared bloodsample, incubate in 1% formaldehyde in D-PBS (30 μl), andprocess as outlined below for sheared samples.

4. Expose the sample to well-defined shear levels (typically4,000 s-1 in the absence of a platelet antagonist to inducesignificant platelet aggregation) in the presence or absenceof a platelet antagonist for prescribed periods of time (typi-cally 30–60 s).

5. Remove a small aliquot (∼3 μl) from the sheared sample andincubate immediately in 1% formaldehyde in D-PBS (30 μl).

6. Incubate the fixed blood samples (pre-sheared and sheared)with a saturating concentration of a fluorescent-labeledplatelet-specific antibody, such as anti-GPIb (6D1)-FITC,for 30 min in the dark.

7. Dilute specimens with 2 ml of 1% formaldehyde, and analyzeby flow cytometry

8. Flow cytometric analysis is used to distinguish platelets fromother blood cells on the basis of their characteristic forward-scatter and fluorescence profiles (Fig. 1.1). Data acquisitionis carried out on each sample for a defined period (usu-ally 100 s), allowing equal volumes for the pre-sheared andsheared specimens to be achieved. Percent platelet aggre-gation is determined based on the disappearance of singleplatelets and increase in platelet aggregates using the follow-ing formula:

% platelet aggregation = (1 − Ns/Nc × 100)

where Ns represent the single platelet population of the shearedspecimen and Nc represents the single platelet population ofthe pre-sheared specimen. By comparing the extent of plateletaggregation in the presence and absence of a platelet antagonist,antiplatelet effects can be readily determined.

3.2. PlateletAdhesion andAggregation UnderDynamic Shear


This section describes an in vitro flow model of platelet throm-bus formation that can be used to evaluate the ex vivo and/orin vitro efficacy of platelet antagonists. Thrombus formationcan be initiated by platelet adhesion from rapidly flowing blood

20 Mousa

Fig. 1.1. Quantification of shear-induced platelet aggregation by flow cytometry. PanelA corresponds to an un-sheared blood specimen. Panel B corresponds to a blood spec-imen that has been subjected to a pathologically high level of shear stress for 30 s. Ascan be seen in the Figure there are three distinct cell populations. The upper populationconsists of platelets and platelet aggregates. The “red blood cell–platelet” populationcorresponds to platelets associated with erythrocytes and leukocytes (see Evaluation,Comment 4). The white blood cell population consists of some leukocytes that haveelevated levels of FITC autofluorescence. The left vertical line separates single platelets(≤ 4.5 μm in diameter) from platelet aggregates, whereas the right vertical line sepa-rates “small” from “large” platelet aggregates. The latter were defined to be larger than10 μm in equivalent sphere diameter.

onto exposed sub-endothelial surfaces of injured vessels pre-senting collagen and vWF, resulting in platelet activation andaggregation. Konstantopolous et al. (43) described the use of aparallel-plate flow chamber that provides a controlled and well-defined flow environment (i.e., chamber geometry and flow ratethrough the chamber). Wall shear stress, τw, assuming a Newto-nian and incompressible fluid, can be calculated using the follow-ing formula:

τw = 6μQwh2 ,

where Q is the volumetric flow rate, μ is the viscosity of the flow-ing fluid, h is the channel height, and w is the channel width.


A typical flow chamber consists of a transparent polycarbonateblock, a gasket whose thickness determines the channel depth,and a glass coverslip coated with an extracellular matrix protein,such as type I fibrillar collagen. The apparatus is held togetherby vacuum. Shear stress is generated by flowing fluid (i.e., anti-coagulated whole blood or isolated cell suspensions) through thechamber over the immobilized substrate under controlled kine-matic conditions using a syringe pump. Recently, Mousa et al.(44) combined the parallel-plate flow chamber with a comput-erized epifluorescence videomicroscopy system, enabling sepa-ration and real-time visualization of adhesion and subsequentaggregation of human platelets in anticoagulated whole blood(or isolated platelet suspensions) flowing over an immobilizedsubstrate.

3.2.2. Procedure

3.2.2.1. Preparationof Collagen-CoatedSurfaces

1. Dissolve 500 mg of collagen type I from bovine achilles ten-don into 200 ml of 0.5 M acetic acid, pH 2.8.

2. Homogenize for 3 h.3. Centrifuge the homogenate at 200×g for 10 min, collect

the supernatant, and then measure collagen concentrationby modified Lowry analysis.

4. Coat glass coverslips with 200 μl of fibrillar collagen I sus-pension so that all but the first 10 mm of the slide is covered(coated area = 12.7 mm×23 mm) and then place coatedcoverslips in a humid environment at 37◦C for 45 min.

5. Rinse the slides to remove excess collagen with 10 ml ofwarm (37◦C) D-PBS and then assembly the flow chamber.

3.2.2.2. PlateletPerfusion

1. Add the fluorescent dye quinacrine dihydrochloride to anti-coagulated whole blood samples at a final concentration of10 μM immediately after blood collection.

2. Prior to the perfusion experiment, incubate blood withplatelet antagonist, or vehicle as a control, at 37◦C for10 min.

3. Perfuse anticoagulated whole blood through the flow cham-ber for 1 min at wall shear rates of 100/s (typical ofvenous circulation) to 1,500/s (representative of partiallyconstricted arteries) for prescribed periods of time (i.e.,1 min). Platelet–substrate interactions are monitored in realtime using an inverted microscope equipped with an epi-fluorescence illumination apparatus and a silicon-intensifiedtarget video camera and recorded on videotape. The micro-scope stage and flow chamber are maintained at 37◦Cby a heating module and incubator enclosure during theexperiment.

22 Mousa

4. Videotaped images are digitized and computer analyzed at5, 15, and 60 s for each perfusion experiment. The numberof adherent individual platelets in the microscopic field ofview during the initial 15 s of flow is determined by imageprocessing and used as the measurement of platelet adhesionthat initiates platelet thrombus formation. The number ofplatelets in each individual thrombus is calculated as the totalthrombus intensity (area×fluorescence intensity) divided bythe average intensity of single platelets determined in the 5-simages. By comparing the extents of platelet aggregation inthe presence and absence of a platelet antagonist, antiplateletefficacy can be determined (Fig. 1.2). Along these lines, anypotential inhibitory effects of a platelet antagonist on plateletadhesion can be readily assessed.

Fig. 1.2. Three-dimensional computer-generated representation of platelet adhesionand subsequent aggregation on collagen I/von Willebrand factor from normal hep-arinized blood perfused in the absence (control) or presence of a GPIIb/IIIa antagonist(XV454) at 37◦C for 1 min at 1,500/s.

3.3. Cell Adhesionto ImmobilizedPlatelets:Parallel-Plate FlowChamber


This section outlines an in vitro flow assay to distinguish agentswith markedly distinct affinities and off-rates. In this assay, immo-bilized platelets are pretreated with a GPIIb/IIIa antagonist,and any unbound drug is washed away before the perfusionof monocytic THP-1 cells. Using this technique, Albulenciaet al. (45) demonstrated that agents with slow platelet off-rates,such as XV454 (t1/2 of dissociation = 110 min; Kd = 1 nM)and abciximab (t1/2 of dissociation = 40 min; Kd = 9.0 nM),which are present predominantly as receptor-bound entities inplasma with little unbound agent, can effectively block plateletheterotypic interactions. In contrast, agents with relatively fast


platelet dissociation rates, such as orbofiban (t1/2 of dissocia-tion = 0.2 min; Kd >110 nM), the antiplatelet efficacy of whichdepends on the plasma concentration of the active drug, do notexhibit such inhibitory effects (44).

3.3.2. Procedure

3.3.2.1. Preparationof 3-Aminopropyltrietho-xysilane (APES)-TreatedGlass Slides

1. Soak glass coverslips overnight in 70% nitric acid.2. Wash coverslips with tap water for 4 h.3. Dry coverslips by washing once with acetone, and then

immerse in a 4% solution of APES in acetone for 2 min.4. Repeat step 3, followed by a final rinse of the glass coverslips

with acetone.5. Wash coverslips 3 times with water and dry overnight.

3.3.2.2. Immobilizationof Platelets onAPES-Treated GlassSlides

1. Layer WP or PRP (2×108 cells/ml) on the surface ofan APES-coated coverslip at a density of approximately30 μl/cm2.

2. Allow platelets to bind to the coverslip in a humid environ-ment at 37◦C for 30 min.

3.3.2.3. MonocyticTHP-1 Cell–PlateletAdhesion Assay

1. Assemble the platelet-coated coverslip in a parallel-plate flowchamber. Mount the chamber on the stage of an invertedmicroscope equipped with a CCD camera connected to aVCR and TV monitor.

2. Perfuse antiplatelet antagonist at the desired concentration,or vehicle as a control, over surface-bound platelets andincubate for 10 min. The extent of platelet activation canbe further modulated by chemical agonists such as thrombin(0.02–2 U/ml) during the 10-min incubation period. Themicroscope stage and flow chamber are maintained at 37◦Cwith a heating module and incubator enclosure during theexperiment.

3. In some experiments, unbound platelet antagonist isremoved by a brief washing step (4 min) prior to the per-fusion of the cells of interest over the platelet layer. Alter-natively, platelet antagonist at the desired concentration iscontinuously maintained in the perfusion buffer during theentire course of the experiment.

4. Perfuse cells (i.e., THP-1 monocytic cells, leukocytes, tumorcells, protein-coated beads) over surface-bound platelets inthe presence or absence of platelet antagonist (see above)at the desired flow rate for prescribed periods of time. Cellbinding to immobilized platelets is monitored in real timeand recorded on videotape.

5. Determine the extent of cell tethering, rolling, and station-ary adhesion to immobilized platelets, as well as the average

24 Mousa

velocity of rolling cells. Comparison of the extent of celltethering, rolling, and stationary adhesion to immobilizedplatelets in the presence and absence of platelet antagonist(Fig. 1.3) will provide a measure of antiplatelet activity.

Fig. 1.3. Phase-contrast photomicrograph of THP-1 cells (phase bright objects) attachedto a layer of thrombin-treated platelets (phase dark objects) after THP-1 cell perfusionfor 3 min at a shear stress level of 1.5 dyn/cm2.

3.3.3. Evaluation 1. Low-speed centrifugation of whole blood results in the sep-aration of platelets (top layer) from larger and denser cellssuch as leukocytes and erythrocytes (bottom layer). To min-imize leukocyte contamination of PRP, slowly aspirate theuppermost 2/3 of the platelet layer after centrifugation.Also, certain rare platelet disorders such as Bernard–Souliersyndrome (BSS) are characterized by larger than normalplatelets which must be isolated by allowing whole bloodto separate by gravity for 2-h post-venipuncture.

2. The mechanical force most relevant to platelet-mediatedthrombosis is shear stress. Normal time-averaged levels ofvenous and arterial shear stress range from 1–5 dyn/cm2

to 6–40 dyn/cm2, respectively. However, fluid shear stressmay reach levels well over 200 dyn/cm2 in small arteriesand arterioles that are partially obstructed by atheroscleroticlesions or vascular spasm. The cone-and-plate viscometerand parallel-plate flow chamber are two of the most com-mon devices used to simulate fluid mechanical shear stress inblood vessels.

3. Due to the high concentration of platelets and erythrocytesin whole blood, small aliquots (3 μl) of pre-sheared andpost-sheared samples must be obtained and processed priorto the flow cytometric analysis. This will minimize artifacts


produced when platelets and erythrocytes pass through thelight beam of a flow cytometer at the same time.

4. The “red blood cell–platelet” population in the flow cytom-etry histogram of Fig. 1.1 represents 3–5% of the dis-played cells. A small fraction (5%) of this population likelyrepresents leukocyte–platelet aggregates, based on analysisusing an anti-CD45 monoclonal antibody. The remainingevents correspond to platelet-associated erythrocytes. How-ever, there is evidence that the majority of the latter popu-lation is an artifact generated by the simultaneous passageof a platelet and an erythrocyte through the light beam ofthe flow cytometer. That this population represents an arti-fact is supported by the observation that further dilution ofpre-sheared and sheared blood specimens and/or reductionof the sample flow rate during the flow cytometric analysisresults in a dramatic decrease in the “red blood cell–platelet”population.

5. Collagen density on the glass coverslip after rinsing withD-PBS can be determined by measuring the difference inaverage weight of 20 clean uncoated slides and 20 collagen-treated slides.

6. Experiments are optimally monitored approximately 100–200 μm downstream of the collagen/glass interface using a60× FLUOR objective and a 1× projection lens, which givesa 3.2×104 μm2 field of view. Monitoring closer to the inter-face may yield non-reproducible results due to variations inthe collagen layering in that region. Positions farther down-stream are to be avoided as well so as to minimize the effectsof upstream platelet adhesion and subsequent aggregationon the fluid dynamic environment and bulk platelet concen-tration.

7. The fluorescence intensity emitted by a single platelet canbe determined by subtracting a digitized background imagetaken at the onset of perfusion, prior to platelet adhesionto the collagen I surface, from a subsequent image acquired5 s after the initiation of platelet adhesion. The fluorescenceintensity of a single platelet is represented by the mean graylevel (black = 0; white = 255) of the platelet, obtained usingimage processing software (i.e., OPTIMAS; Agris-SchoenVision Systems, Alexandria, VA, USA), multiplied by thecorresponding area (total number of pixels) covered by theplatelet. The intensity values for all single platelet events areaveraged at the 5-s time point to arrive at an average singleplatelet intensity.

8. A single field of view (10×; 0.55 mm2) is monitored dur-ing the 3-min period of the experiment. At the conclusion

26 Mousa

of the experiment, five additional fields of view (0.55 mm2)are monitored for 15 s each. The following parameters canbe quantified: (a) the total number of interacting cells permm2 during the 3-min perfusion experiment; (b) the num-ber of stationary interacting cells per mm2 after 3 min ofshear flow; (c) the percentage of total interacting cells thatare stationary after 3 min of shear flow; and (d) the averagerolling velocity (μm/s) of interacting cells. The number ofinteracting cells/mm2 is determined manually by reviewingthe videotapes. Stationary interacting cells/mm2 are thosecells that move <1-cell radius within 10 s of the conclusionof the 3-min perfusion. To quantify cell numbers, images canbe digitized from the videotape recorder using an imagingsoftware package (e.g., OPTIMAS). Rolling velocities canbe computed using image processing as the distance trav-eled by the centroid of the rolling cell divided by the timeinterval.

References

1. Konstantopoulos, K., Kukreti, S., and McIn-tire, L.V. (1998) Biomechanics of cell inter-actions in shear fields Adv drug delivery rev33, 141–64.

2. Alevriadou, B.R., Moake, J.L., Turner, N.A.,Ruggeri, Z.M., Folie, B.J., Phillips, M.D.,Schreiber, A.B., Hrinda, M.E., and McIn-tire, L.V. (1993) Real-time analysis ofshear-dependent thrombus formation and itsblockade by inhibitors of von Willebrand fac-tor binding to platelets Blood 81, 1263–76.

3. Turitto, V.T. (1982) Blood viscosity, masstransport, and thrombogenesis Prog HemostThromb 6, 139–77.

4. Palmer, R.L. (1984) Laboratory diagnosis ofbleeding disorders. Basic screening tests Post-graduate Med 76, 137–42, 47–8.

5. Nilsson, I. Assessment of blood coagulationand general haemostasis. In: Bloom, A.L, andThomas, D.P., eds. Haemostasis and Throm-bosis, 2nd edition. Harlow: Longman Group(UK) Limited;1987:922–32

6. Hartert, H. (1948) Blutgerinnungsstudienmit der Thrombelastographie, einem neuenUntersuchungsverfahren Klinische Wochen-schrift 26, 577–83.

7. Khurana, S., Mattson, J.C., Westley, S.,O’Neill, W.W., Timmis, G.C., and Safian,R.D. (1997) Monitoring platelet glycopro-tein IIb/IIIa-fibrin interaction with tissuefactor-activated thromboelastography J LabClin Med 130, 401–11.

8. Mousa, S., Khurana, S., and Forsythe, M.(2000) Comparative in vitro efficacy of dif-

ferent platelet glycoprotein IIb/IIIa antag-onists on platelet-mediated clot strengthinduced by tissue factor with use of throm-boelastography: differentiation among gly-coprotein IIb/IIIa antagonists ArteriosclerThromb Vasc Biol 20, 1162–7.

9. Bhargava, A.S., Freihube, G., and Gunzel,P. (1980) Characterization of a new potentheparin. 3rd Communication: determina-tions of anticoagulant activity of a newpotent heparin preparation by thrombelas-tography in vitro using citrated dog andhuman blood Arzneimittel-Forschung 30,1256–8.

10. Barabas, E., Szell, E., and Bajusz, S. (1993)Screening for fibrinolysis inhibitory effect ofsynthetic thrombin inhibitors Blood CoagulFibrinol 4, 243–8.

11. Scherer, R.U., Giebler, R.M., Schmidt, U.,Paar, D., Wust, T., Spangenberg, P., Militzer,K., Hirche, H., and Kox, W.J. (1995) Short-time rabbit model of endotoxin-inducedhypercoagulability Lab Animal Sci 45,538–46.

12. Mousa, S.A., Bozarth, J.M., Seiffert, D.,and Feuerstein, G.Z. (2005) Using thrombe-lastography to determine the efficacy ofthe platelet glycoprotein IIb/IIIa antago-nist, roxifiban, on platelet/fibrin-mediatedclot dynamics in humans Blood Coagul Fib-rinol 16, 165–71.

13. Zuckerman, L., Cohen, E., Vagher, J.P.,Woodward, E., and Caprini, J.A. (1981)Comparison of thrombelastography with


common coagulation tests Thromb Haemost46, 752–6.

14. Chandler, A.B. (1958) In vitro thromboticcoagulation of the blood; a method for pro-ducing a thrombus Lab Invest; J Tech MethodsPathol 7:110–4.

15. Robbie, L.A., Young, S.P., Bennett, B., andBooth, N.A. (1997) Thrombi formed in aChandler loop mimic human arterial thrombiin structure and RAI-1 content and distribu-tion Thromb Haemost 77, 510–5.

16. Stringer, H.A., van Swieten, P., Heijnen,H.F., Sixma, J.J., and Pannekoek, H. (1994)Plasminogen activator inhibitor-1 releasedfrom activated platelets plays a key rolein thrombolysis resistance. Studies withthrombi generated in the Chandler loopArterioscler Thromb 14, 1452–8.

17. van Giezen, J.J., Nerme, V., and Abrahams-son, T. (1998) PAI-1 inhibition enhances thelysis of the platelet-rich part of arterial-likethrombi formed in vitro. A comparative studyusing thrombi prepared from rat and humanblood Blood Coagul Fibrinol 9, 11–8.

18. Born, G. (1962) Quantitative investigationsinto the aggregation of blood platelets J Phys-iol (London) 162, 67P–8P.

19. Born, G.V. (1962) Aggregation of bloodplatelets by adenosine diphosphate and itsreversal Nature 194, 927–9.

20. Haskel, E.J., and Abendschein, D.R. (1989)Deaggregation of human platelets in vitro byan RGD analog antagonist of platelet glyco-protein IIb/IIIa receptors Thromb Res 56,687–95.

21. Breddin, K., Grun, H., Krzywanek, H.J.,and Schremmer, W.P. (1975) Measurementof spontaneous platelet aggregation. Plateletaggregation test III (author’s transl) Klinis-che Wochenschrift 53, 81–9.

22. Klose, H.J., Rieger, H., and Schmid-Schonbein, H. (1975) A rheological methodfor the quantification of platelet aggregation(PA) in vitro and its kinetics under definedflow conditions Thromb Res 7, 261–72.

23. Marguerie, G.A., Edgington, T.S., and Plow,E.F. (1980) Interaction of fibrinogen with itsplatelet receptor as part of a multistep reac-tion in ADP-induced platelet aggregation JBiol Chem 255, 154–61.

24. Marguerie, G.A., Plow, E.F., and Edging-ton, T.S. (1979) Human platelets possess aninducible and saturable receptor specific forfibrinogen J Biol Chem 254, 5357–63.

25. Lumley, P., and Humphrey, P.P. (1981) Amethod for quantitating platelet aggregationand analyzing drug-receptor interactions onplatelets in whole blood in vitro J PharmacolMethods 6, 153–66.

26. Fratantoni, J.C., and Poindexter, B.J.(1990) Measuring platelet aggregation withmicroplate reader. A new technical approachto platelet aggregation studies Am J ClinPathol 94, 613–7.

27. Ammit, A.J., and O’Neill, C. (1991)Rapid and selective measurement of platelet-activating factor using a quantitative bioassayof platelet aggregation J Pharmacol Methods26, 7–21.

28. Markell, M.S., Fernandez, J., Naik, U.P.,Ehrlich, Y., and Kornecki, E. (1993)Effects of cyclosporine-A and cyclosporine-G on ADP-stimulated aggregation of humanplatelets Ann N Y Acad Sci 696, 404–7.

29. Cardinal, D.C., and Flower, R.J. (1980) Theelectronic aggregometer: a novel device forassessing platelet behavior in blood J Phar-macol Methods 3, 135–58.

30. Tohgi, H., Takahashi, H., Watanabe, K.,Kuki, H., and Shirasawa, Y. (1996) Devel-opment of large platelet aggregates fromsmall aggregates as determined by laser-lightscattering: effects of aggregant concentrationand antiplatelet medication Thromb Haemost75, 838–43.

31. Matsuo, T., Koide, M., Sakuramoto, H.,Matsuo, M., and Fuji, K. (1997) Small sizeof platelets aggregates detected by light scat-tering in young habitual smokers ThrombHaemost 71(suppl), 71.

32. Xiao, Z., and Theroux, P. (1998) Plateletactivation with unfractionated heparin attherapeutic concentrations and comparisonswith a low-molecular-weight heparin andwith a direct thrombin inhibitor Circulation97, 251–6.

33. Bennett, J.S., and Vilaire, G. (1979) Expo-sure of platelet fibrinogen receptors by ADPand epinephrine J Clin Invest 64, 1393–401.

34. Mendelsohn, M.E., O’Neill, S., George, D.,and Loscalzo, J. (1990) Inhibition of fib-rinogen binding to human platelets by S-nitroso-N-acetylcysteine J Biol Chem 265,19028–34.

35. Gallimore, M.J., Tyler, H.M., and Shaw,J.T. (1971) The measurement of fibrinoly-sis in the rat Thromb Diath Haemorrh 26,295–310.

36. Teitel, P. (1977) Basic principles of the ‘Fil-terability test’ (FT) and analysis of erythro-cyte flow behavior Blood Cells 3, 55–70.

37. Kiesewetter, H., Dauer, U., Teitel, P.,Schmid-Schonbein, H., and Trapp, R.(1982) The single erythrocyte rigidometer(SER) as a reference for RBC deformabilityBiorheology 19, 737–53.

38. Roggenkamp, H.G., Jung, F., and Kiesewet-ter, H. (1983) A device for the electrical

28 Mousa

measurement of the deformability of redblood cells Biomedizinische Technik 28,100–4.

39. Seiffge, D., and Behr, S. (1986) Passage ofred blood cells in the SER: their distributionand influences of various extrinsic and intrin-sic factors Clin Hemorheol 6, 1510164.

40. Schmid-Schonbein, H., von Gosen, J.,Heinich, L., Klose, H.J., and Volger, E.(1973) A counter-rotating "rheoscope cham-ber" for the study of the microrheology ofblood cell aggregation by microscopic obser-vation and microphotometry Microvasc Res6, 366–76.

41. Jen, C.J., and McIntire, L.V. (1984)Characteristics of shear-induced aggregationin whole blood J Lab Clin Med 103,115–24.

42. Sweeney, J.D., Labuzzetta, J.W., Michielson,C.E., and Fitzpatrick, J.E. (1989) Wholeblood aggregation using impedance and par-ticle counter methods Am J Clin Pathol 92,794–7.

43. Konstantopoulos, K., Kamat, S.G., Schafer,A.I., Banez, E.I., Jordan, R., Kleiman, N.S.,and Hellums, J.D. (1995) Shear-inducedplatelet aggregation is inhibited by in vivoinfusion of an anti-glycoprotein IIb/IIIaantibody fragment, c7E3 Fab, in patientsundergoing coronary angioplasty Circulation91, 1427–31.

44. Mousa, S.A., Abulencia, J.P., McCarty, O.J.,Turner, N.A., and Konstantopoulos, K.(2002) Comparative efficacy between theglycoprotein IIb/IIIa antagonists roxifibanand orbofiban in inhibiting platelet responsesin flow models of thrombosis J CardiovascPharmacol 39, 552–60.

45. Abulencia, J.P., Tien, N., McCarty, O.J.,Plymire, D., Mousa, S.A., and Konstan-topoulos, K. (2001) Comparative antiplateletefficacy of a novel, nonpeptide GPIIb/IIIaantagonist (XV454) and abciximab (c7E3) inflow models of thrombosis Arterioscl ThrombVasc Biol 21, 149–56.

Chapter 2

In Vivo Models for the Evaluation of Antithromboticsand Thrombolytics

Shaker A. Mousa

Abstract

The development and application of animal models of thrombosis have played a crucial role in thediscovery and validation of novel drug targets and the selection of new agents for clinical evaluation,and have informed dosing and safety information for clinical trials. These models also provide valu-able information about the mechanisms of action/interaction of new antithrombotic agents. Small andlarge animal models of thrombosis and their role in the discovery and development of novel agents aredescribed. Methods and major issues regarding the use of animal models of thrombosis, such as posi-tive controls, appropriate pharmacodynamic markers of activity, safety evaluation, species specificity, andpharmacokinetics, are highlighted. Finally, the use of genetic models of thrombosis/hemostasis and howthese models have aided in the development of therapies that are presently being evaluated clinically arepresented.

Key words: Animal models, coronary thrombosis, antithrombotic agents, pharmacodynamics,Folts, Wessler, thrombocytopenia, genetic models.

1. Introduction

The general understanding of the pathophysiology of thrombosisis based on the observations of Virchow in 1856. Three factorsresponsible for thrombogenesis are proposed including obstruc-tion of blood flow, changes in the properties of blood constituents(hypercoagulability), and vessel wall injury. Experimental modelsof thrombosis focus on one, two, or all three factors of Virchow’striad. Therefore, they differ with respect to the prothromboticchallenge, i.e., stenosis, stasis, vessel wall injury (mechanical, elec-trical, chemical, photochemical, laser-light), insertion of foreign


29

30 Mousa

surface, or injection of a prothrombotic factor, as well as vesseltype and animal species.

Roughly, two types of models can be differentiated (1): (1)models in which thrombi are produced in veins by stasis and/orinjection of a procoagulant factor resulting in fibrin-rich “red”venous type thrombi and (2) models in which thrombi are pro-duced in arteries by vessel wall injury and/or stenosis resultingin platelet-rich “white” mural thrombi. This distinction is notstrict because platelets and the coagulation system influence eachother. Drugs that prevent fibrin formation may well act in arterialmodels and vice versa. Thrombosis models are usually performedin healthy animals. The underlying chronic diseases in human,namely atherosclerosis or thrombophilias, are not addressed inthe models. Thus, any model is limited with regard to clinical rele-vance. The pharmacological effectiveness of a new antithromboticdrug should be studied in more than one animal model. Despitethese limitations, animal models predict the clinical effectivenessof drugs for the treatment and prevention of thrombotic diseasesfairly well. A list of such drugs is presented in a recent review byLeadleey et al. (2). The clinical usefulness of an antithromboticdrug is determined in part by safety/efficacy ratio with respectto bleeding risk. Assessment of this parameter of the hemostaticsystem should therefore be included in the models if possible.

The development of antithrombotic agents requires preclin-ical assessment of the biochemical and pharmacologic effects ofthese drugs. It is important to note that second- and third-generation antithrombotic drugs are devoid of in vitro antico-agulant effects, yet in vivo, by virtue of endogenous interactions,these drugs produce potent antithrombotic actions. The initialbelief that an antithrombotic drug must exhibit in vitro anticoag-ulant activity is no longer valid. This important scientific obser-vation has been possible only because of the availability of animalmodels.

Several animal models utilizing species such as rats, rabbits,dogs, pigs, and monkeys have been made available for routine use.Other animal species such as the hamster, mouse, cat, and guineapig have also been used. Species variations are an important con-sideration in selecting a model and interpreting the results, asthese variations can result in different antithrombotic effects.Rats and rabbits are the most commonly used species in whichboth arterial and venous thromboses have been investigated. Bothpharmacologic and mechanical means have been used to producea thrombogenic effect in these models. Both rat and rabbit mod-els for studying bleeding effects of drugs have also been devel-oped. The rabbit ear blood loss model is most commonly used totest the hemorrhagic effect of drugs. The rat tail bleeding mod-els have also been utilized for the study of several antithromboticdrugs.

These animal models have been well established and can beused for the development of antithrombotic drugs. It is also pos-

In Vivo Models for the Evaluation of Antithrombotics and Thrombolytics 31

sible to use the standardized bleeding and thrombosis modelsto predict the safety and efficacy of drugs. Thus, in additionto the evaluation of in vitro potency, the endogenous effectsof antithrombotic drugs can be investigated. Such standardizedmethods can be recommended for inclusion in pharmacopoeialscreening procedures. Numerous models have now been devel-oped to mimic a variety of clinical conditions where antiplateletand antithrombotic drugs are used, including myocardial infarc-tion, stroke, cardiopulmonary bypass, trauma, peripheral vasculardiseases, and restenosis. While dog and primate models are rel-atively expensive, they have also provided useful information onthe pharmacokinetics and pharmacodynamics of antithromboticdrugs. The primate models in particular have been extremelyuseful, as the hemostatic pathways in these species are com-parable to those in humans. The development of such agentsas the specific glycoprotein IIb/IIIa inhibitor antibodies relieslargely on these models. These models are of pivotal value in thedevelopment of antithrombotic drugs and provide extremely use-ful data on the safety and efficacy of new drugs developed forhuman use.

In most animal models of thrombosis, healthy animals arechallenged with thrombogenic (pathophysiologic) stimuli and/orphysical stimuli to produce thrombotic or occlusive conditions.These models are useful for the screening of antithromboticdrugs.

1.1. StasisThrombosis Model

Since its introduction by Wessler in 1959 (3), the rabbit model ofjugular stasis thrombosis has been extensively used for the phar-macologic screening of antithrombotic agents. This model hasalso been adapted for use in rats (4). In the stasis thrombosismodel, a hypercoagulable state is mimicked by the administrationof one of a number of thrombogenic challenges, including humanserum (5), thromboplastin (6), activated prothrombin complexconcentrates (7), factor Xa (8) and recombinant relipidated tis-sue factor (9). Administration of such agents produces a hyper-coagulable state. Diminution of blood flow achieved by ligatingthe ends of vessel segments serves to augment the prothromboticenvironment. The thrombogenic environment produced in thismodel simulates venous thrombosis where both the blood flowand the activation of coagulation play a role in the developmentof a thrombus.

1.2. Models Based onVessel Wall Damage

The formation of a thrombus is not solely induced by a plas-matic hypercoagulable state. In the normal vasculature, the intactendothelium provides a non-thrombogenic surface over whichthe blood flows. Disruption of the endothelium exposes sub-endothelial tissue factor and collagen, which activate the coagula-tion and platelet aggregation processes, respectively. Endothelialdamage can be induced experimentally by physical means

32 Mousa

(clamping, catheter), chemical means (fluorescein isothyocynate,Rose Bengal, ferrous chloride), thermal injury, or electrolyticinjury.

Each setting in the design of an animal model can answerspecific question in relation to certain thrombotic disorders inhuman. However, the ultimate model of human thrombosis is inhuman.

1.3. Issues to beTaken inConsideration inEvaluatingAntithrombotics

1.3.1. Effect onHemostasis

Antithrombotic and anticoagulant drugs are effective in the con-trol of thrombogenesis at various levels. These drugs are alsocapable of producing hemorrhagic effects that cannot be pre-dicted using in vitro testing methods. The bleeding effects of adrug may be direct or indirect; hence assessing efficacy/safetyratios could be a useful parameter.

Single versus repeated exposure. Repeated administration ofdrugs can result in a cumulative response that may alter the phar-macokinetic and pharmacodynamic indices of a given agent. Itis only through the use of animal models that such informationcan be generated. Furthermore, since antithrombotic drugs rep-resent a diverse class of agents, their interactions with physiologi-cally active endogenous proteins can only be studied using animalmodels.

1.3.2. Choice of Species Species variation plays an important role in thrombotic, hemo-static, and hemorrhagic responses. While there is no set formulato determine the relevance of the results obtained with animalmodels to man, the use of animal models can provide valuableinformation on the relative potency of drugs, their bioavailabilityafter various routes of administration, and their pharmacokineticbehavior. Specific studies have provided data on the species rele-vance of the responses in different animal models to the projectedhuman responses. Thus, the use of animal models in the evalua-tion of different drugs can provide useful data to compare differ-ent drugs within a class. However, caution must be exercised inextrapolating such results to the human clinical condition.

1.3.3. Selection ofAnimal Model

The selection of animal model for the evaluation of antithrom-botic effects depends on several factors. The interaction of a givendrug with the blood and vascular components and its metabolictransformation are important considerations. Thus, ex vivo anal-ysis of blood along with the other endpoints can provide usefulinformation on the effects of different drugs. Unlike the screeningof drugs such as antibiotics, antithrombotic drugs require multi-


parametric endpoint analysis. Animal models are the most usefulsystem in the evaluation of the effects of these drugs.

Finally, it should be stressed that the evaluation of pharma-copoeial and in vitro potency of antithrombotic drugs does notnecessarily reflect the in vivo safety/efficacy profile. Endogenousmodulation, such as the release of tissue factor pathway inhibitor(TFPI) by heparins, plays a very important role in the overall ther-apeutic index of many drugs. Such data can only be obtainedusing animal models. It is therefore important to design exper-iments where several data points can be obtained. This informa-tion is of crucial value in the assessment of antithrombotic drugsand cannot be substituted by other in vitro or tissue culture-basedmethods.

Several excellent reviews covering the different theoreticaland technical aspects of thrombosis and thrombolysis modelshave been published previously (2, 11–15). These reviews, alongwith more specialized reviews of models of atherosclerosis (16),restenosis (17), and stroke (18) provide comprehensive informa-tion regarding the details of many models and provide the patho-logical rationale for using specific models for specific diseases.In addition, the advantages and disadvantages of each model ofthrombosis and thrombolysis are described in these reviews. Thischapter focuses on the use of thrombosis models in the drug dis-covery process, with emphasis on the practical application of thesemodels. Examples from studies evaluating therapeutic approachesthat target various antithrombotic mechanisms will be presentedto demonstrate the current use of thrombosis models in drug dis-covery. Important issues in evaluating novel antithrombotic com-pounds will also be addressed. In addition, evidence demonstrat-ing the clinical relevance of preclinical data derived from animalmodels of thrombosis will also be presented. Finally, a summaryof the use of genetic models of thrombosis/hemostasis and theircurrent and potential use in drug discovery will also be discussed

2. Animal Modelsof Thrombosis

2.1. Stenosis- andMechanicalInjury-InducedCoronary Thrombosis(Folts Model)

2.1.1. Purposeand Rationale

Thrombosis in stenosed human coronary arteries is one of themost common thrombotic diseases leading to unstable angina,

34 Mousa

Fig. 2.1. Technique for monitoring platelet aggregation in the partially obstructed leftcircumflex coronary artery of the dog. Electromagnetic flow probes measure blood flowiii nil/mm. Partial obstruction of the coronary artery with a plastic Lexan cylinder resultsin episodic cyclical reductions in coronary blood flow that are due to platelet-dependentthrombus formation. Every 2–3 mm the thrombus must be mechanically shaken loose(SL) to restore blood.

acute myocardial infarction, or sudden death. Treatment withangioplasty, thrombolysis, or bypass grafts can expose new throm-bogenic surfaces, and re-thrombosis may occur. The mechanismsresponsible for this process include interactions of platelets withthe damaged arterial wall and platelet aggregation.

In 1976, Folts and co-workers (19) described a model ofrepetitive thrombus formation, or cyclic flow reductions (CFRs),in stenosed coronary arteries of open-chest, anesthetized dogs(Fig. 2.1). This model is also applicable to the rabbit femoralor carotid artery (20). Using this model, several groups havedescribed the antithrombotic effects of a variety of drugs, primar-ily prostaglandin-inhibitors, prostacyclin-mimetics, or fibrinogenreceptor antagonists (21–27). The combination of two thrombo-genic stimuli leads to the development of CFRs in this model:severe, concentric stenosis and focal, intimal injury. With fewexceptions, CFRs will not develop unless both stimuli exist.


The rheological conditions required to produce turbulenceand stasis upon vessel narrowing dictate that lumenal diameter bereduced by at least 50%. Two- to three-millimeter long constric-tors are cut from Lexan R© rods readily available from local plasticsdistributors. One center hole of varying diameter and two smallercollar holes are drilled, into which the prongs of snap-ring pliersfit to spread the constrictor’s slit in the top-central portion toapply or remove them. Other plastics will suffice, but Lexan isideal because of its strength and resiliency. Both circumflex andleft anterior descending (LAD) coronary arteries have been usedin this model. Besides personal preferences, we know of no phys-iologic basis for preferring one to the other.

Owing to the prominent auto-regulation of coronary circu-lation, it is difficult to assess the severity of a stenosis on thebasis of changes in basal coronary blood flow (CBF). However,the robust reactive hyperemia (RH) characteristic of the coro-nary circulation provides a powerful tool with which to gauge theseverity of the stenosis. With gradual narrowing, basal CBF willremain unchanged or decline negligibly, whereas RH will beginto decline sooner as the vasodilatory reserve of downstream ves-sels is progressively exhausted. Reduction of lumenal diameter tothis degree is required if one wishes to produce CFRs in a highpercentage (i.e., >90%) of dogs. It is also critical if one wishes tocompare two or more drugs in this model and draw meaningfulconclusions about drug effects. Inasmuch as the severity of thestenosis is an important component of the thrombogenic stim-ulus, comparable and uniform degrees of constriction betweentreatment groups are required, preferably those in which basalCBF is reduced between 10 and 25% and RH is abolished, ornearly so. It is important to apply these criteria when one is inves-tigating a drug that possesses vasodilatory effects or one whosepharmacologic profile is not completely known. Without exhaus-tion of the vasodilatory reserve (as evidenced by abolition of RH),elimination of CFRs could result (at least partly) from coronaryvasodilation. One difficulty in using RH or basal flow reductionimmediately after placing a constrictor on the coronary artery isthat CBF starts to decline quickly as platelets accumulate at thesite of stenosis and intimal injury. Thus, one needs to assess thedegree of flow reduction immediately after constricting the artery.Delaying this assessment will result in an overestimation of thestenosis severity due to accumulation of platelets on the vessellining. Alternatively, the degree of stenosis can be ascertained byapplying the constrictor before denuding the artery (see below), inwhich case the constrictor (or constrictors) needs to be removedand reapplied.

After damaging and stenosing the coronary artery sufficiently,CBF starts declining immediately, reaching zero within 4–12 min,and remaining there until blood flow is restored by manually

36 Mousa

shaking loose the thrombus (“SL,” see Fig. 2.1, bottom).This is usually accomplished by either flicking the Lexan con-strictor or sliding the constrictor up and down the artery tomechanically dislodge the thrombus. Spontaneous flow restora-tions occur under three circumstances: (1) non-severe conditions(i.e., minimal stenosis or de-endothelialization); (2) waning CFRs(which can occur as late as 30–45 mm after establishing CFRs);and/or (3) administration of a partially effective antithromboticagent.

Although the influence of blood pressure on the rate of for-mation of occlusive thrombi or their stability has not been stud-ied systematically, one might predict that higher arterial pressureswould increase the deceleration of CBF to zero by enhancingplatelet aggregation through increased shear forces and increaseddelivery of platelets to the growing thrombus. Higher arterialpressure also might increase the propensity for spontaneous flowrestorations before an occlusive thrombus is formed, due togreater stress on the nascent, unconsolidated thrombus.

Several groups have examined histologically the coronaryarteries harvested from dogs undergoing CFRs usually when CBFis declining or has ceased. Extensive intimal injury, including de-endothelialization with adherent platelets and/or microthrombi,is consistently observed. Arteries harvested when CBF is zeroinvariably reveal a platelet-rich thrombus filling the stenotic seg-ment (19, 21, 23, 28). These histological observations, cou-pled with the pattern of gradual, progressive declines in CBFand abrupt increases thereof (whether spontaneous or deliberate),provide further evidence that CFRs indeed are caused primarily byplatelet thrombi, not vasoconstriction.

Although the primary cause of CFRs is platelet aggrega-tion, it is possible that local vasospasm and/or vasoconstric-tion downstream from the site of thrombosis are induced byvasoactive mediators released by activated and/or aggregat-ing platelets. Experimental evidence supporting vasoconstric-tion just downstream from the stenosis during CFRs has beendemonstrated (29).

Further evidence for platelet-dependent thrombus formationin the etiology of CFRs is derived from the pharmacologicalprofile of this model. In general, platelet-inhibitory agents con-sistently abolish or attenuate CFRs, whereas vasodilators (e.g.,nitroglycerin, calcium entry blockers, and papaverine) affect themnegligibly (30). Aspirin was the first described inhibitor of CFRs(21). However, in subsequent studies, its effects on CFRs werefound to be variable and dose-dependent (22). Variability in theresponse to aspirin may be related to the severity of the stenosis,as further tightening of the constrictor after an effective dose ofaspirin or ibuprofen usually restores CFRs.

Prostacyclin, a powerful anti-aggregatory and potent coro-nary vasodilatory product of endothelial arachidonic acid


metabolism, is extremely efficacious and potent in abolishingCFRs. It is noteworthy that different drug classes can becompared, as evidenced by the wide range of percentages ofresponders (28).

Advances in platelet physiology and pharmacology have iden-tified a new class of antiplatelet agents that block the plateletmembrane glycoprotein IIb/IIIa (GPIIb/IIIa) receptor andhence fibrinogen binding. Fibrinogen binding between plateletsis an obligate event in aggregation and is initiated by blood-borneplatelet agonists such as ADP, serotonin, thrombin, epinephrine,and collagen (31). The tripeptide sequence Arg–Gly–Asp (RGD),which occurs twice in the Aα-chain of fibrinogen, is believedto mediate, at least in part, the binding of fibrinogen to theGPIIb/IIIa complex.

Early experimental results with GPIIb/IIIa antagonists instudies by Coller et al. (32), Bush (26), and Shebuski (25, 33)demonstrated that fibrinogen receptor antagonists are as effec-tive as prostacyclin as anti-aggregatory and antithrombotic agentsand do not possess the hemodynamic liabilities associated withprostaglandin-based compounds. Monoclonal antibodies directedagainst the platelet fibrinogen receptor (abciximab) are essentiallyirreversible, whereas RGD (tirofiban)- or KGD (eptifibatide)-based fibrinogen receptor antagonists are reversible, their effectsdissipating within hours after discontinuation of intravenousinfusion.

The prominence of platelet aggregation vis-à-vis coagulationmechanisms in the Folts model is evidenced by the lack of effectof heparin and thrombin inhibitors reported by most investigators(19, 22). However, heparin and MCI-9038, a thrombin inhibitor,were reported to abolish CFRs in about two-thirds of dogs withrecently (30 min) established CFRs, but not in those extent after3 h (34). The explanation for the differential effects of heparinis not immediately apparent. It may be related to the severity ofthe stenosis used. These apparently discrepant observations couldbe related to inhibition of thrombin-stimulated platelet activationand/or aggregation.

An attractive feature of the Folts model is its amenability todose-response studies. Unlike other models in which the throm-botic processes are dynamic, occurring over several minutes tohours, CFRs in the Folts model are repetitive and remarkablyunchanging. In the many dogs that received either no interven-tion or vehicle 1 h after initiating CFRs, flow patterns remainedunchanged for at least another hour (23). Thus, one can evalu-ate several doses of an investigational drug in a single dog. Weand others have exploited this to determine potencies, an impor-tant basis of comparison between drugs with similar mechanismsof action, thus underscoring another feature of the model: itsamenability to quantification of drug response. Two methods forquantifying drug effects in this model have been described.

38 Mousa

Aiken et al. (21) first described a four-point scoring scheme toassess and compare different doses or drugs, ranging from 0 (noeffect on CFRs) to 3 (fully effective; complete abolition of CFRs).Intermediate scores of 1 and 2 are respectively applied when theCFR frequency was slowed (but occlusive thrombi still occurred)and when non-occlusive, spontaneously embolizing thrombi wereobserved. An advantage of this system is the provision of a sin-gle number for each evaluation period. A disadvantage is thatagents that decrease systemic blood pressure (e.g., prostacyclin)will also decrease coronary perfusion pressure; the coronary flowpattern will be affected, making the scoring system somewhatmore subjective.

Another method of quantifying CFRs, described by Bushet al. (23), addresses the frequency, expressed on a per hourbasis, and severity, based on the average nadir of CBF before aflow restoration. This system is less subjective, but it producestwo values per evaluation period, and combinations of the two inan effort to provide a single parameter are awkward. In practicalterms, both methods for quantifying CFRs described above pro-vide similar answers. The important point for both is consistencyin scoring. This end is best served by well-defined and communi-cated criteria.

To date, the Folts model has been used only to evaluateantithrombotic drugs. No description of this model for the evalu-ation of thrombolytic drugs or adjunctive agents has been made.However, preliminary data reveal these thrombi to be resistant todoses of thrombolytic agents that lyse thrombi in other models(35). Of all the models described in this review, the thrombi inthis model are probably the most platelet-rich and possess rela-tively less fibrin than, for example, the copper coil or wire mod-els. However, it may be erroneous to conclude that these thrombiare devoid of fibrin, as the fibrinogen that links platelets duringaggregation via the GPIIb/IIIa receptor is theoretically capableof undergoing fibrin formation.

Several investigators have shown that the same combinationof severe vessel narrowing and de-endothelialization results inCFRs in arteries other than the coronary. We have elicited CFRsin femoral arteries in anesthetized dogs with similar degrees ofvessel narrowing and deliberate denudation of the artery (unpub-lished observation). Folts et al. (24) demonstrated that CFRscan be produced in conscious dogs with chronically implantedLexan R© coronary constrictors and flow probes. CFRs were pre-vented in the interim between implantation and acute study bythe administration of aspirin. Al-Wathiqui (36) and Gallagher andco-workers (37) have demonstrated that progressive carotid orcoronary arterial narrowing with ameroid constrictors will resultin CFRs days to weeks after surgical implantation. These dogsapparently did not undergo deliberate vessel denudation at the


time of implantation. Perhaps focal inflammation developed in theintervening week(s) between the surgery and the development ofCFRs in these animals. Alternatively, there was sufficient intimalvessel injury during implantation of the ameroid constrictors toinduce development of CFRs at a later time. CFRs have also beenelicited in the renal (38) and carotid (39) arteries of cynomolgusmonkeys.

Eidt et al. (40) showed that conscious dogs equipped with thesame constrictors over segments of the LAD showing endothe-lial injury undergo repetitive CFRs in response to exercise, butnot ventricular pacing. The frequency and severity of CFRs variedmore in this model, and some CFRs were non-occlusive. CFRsof most dogs eventually deteriorated to persistent no- or low-flow states. Unlike the open-chest preparation, flow restorationsobserved in this model occurred spontaneously. Also, the sever-ity of the stenosis produced was not as great as that produced bymost practitioners of the Folts model, as reflected by the abilityof CBF to increase above control levels initially during exercise.

In summary, the Folts model of platelet-dependent throm-bus formation is a well-established method to determine thepharmacology of antithrombotic agents. It represents an excel-lent choice for initial evaluation of antiplatelet activity in vivo,regardless of the artery used. Qualitatively, the thrombogenicstimuli in the Folts model and those responsible for unstableangina may be similar, since an involvement by platelets has clearlybeen demonstrated in the model and is strongly suspected clini-cally. It should be remembered, however, that flow restorationsin the Folts model require vigorous shaking. In contrast, unsta-ble angina is believed not to involve persistent, total thromboticcoronary occlusion. On the basis of the model’s pharmacologicalprofile, the thrombi in this model also do not appear to resem-ble those usually responsible for acute myocardial infarction, asthe former appear to be unresponsive to thrombolytic agents.The preliminary observations that either streptokinase (SK) ortissue plasminogen activator (t-PA) does not lyse thrombi in theFolts model contrast with the 50–75% response rate to throm-bolytic therapy in patients with evolving myocardial infarction(35). However, it is tempting to speculate that the platelet-richthrombi produced in this model are more like thrombi in thosepatients whose coronary arteries are not reopened by even earlyintervention and/or high doses of t-PA (41), and thus could rep-resent a model of “thrombolytic-resistant” coronary thrombosis.

In order to study new drugs for their antithrombotic poten-tial in coronary arteries, Folts and Rowe (42) developed a modelof periodic acute platelet thrombosis and CFRs in stenosed caninecoronary arteries. Uchida described a similar model in 1975 (27).The model includes various aspects of unstable angina pectoris,i.e., critical stenosis, vascular damage, downstream vasospasm

40 Mousa

induced by vasoconstrictors released or generated by platelets.The cyclic variations in CBF are a result of acute platelet thrombiwhich may occlude the vessel but which either embolize spon-taneously or can easily be embolized by shaking the constrictingplastic cylinder. CFRs are not a result of vasospasm (24). Clini-cally, aspirin can reduce the morbidity and mortality of coronarythrombotic diseases but its effect is limited. Similarly, CFRs in theFolts model are abolished by aspirin but the effect can be reversedby increases in catecholamines and shear forces (43). As part of anexpert meeting on animal models of thrombosis, a review of theFolts model has been published (44).

In this section, five different protocols are described for theinduction of coronary thrombosis.

2.1.1.1. CoronaryThrombosis Induced byStenosis (Protocols 1–4)

The first four protocols are characterized by episodic, sponta-neous decreases in CBF interrupted by restorations of blood flow.These alterations in CBF, or CFRs, are associated with transientplatelet aggregation at the site of the coronary constriction and anabrupt increase in blood flow after embolization of platelet-richthrombi.

Damage to the vessel wall is achieved by placing a hemostaticclamp on the coronary artery. A fixed amount of stenosis is pro-duced by an externally applied obstructive plastic cylinder at thedamaged part of the vessel. In dogs, stenosis is critical; the reac-tive hyperemic response to a 10-second (s) occlusion is abolished(protocol 1). In pigs, stenosis is subcritical; partial reactive hyper-emia remains (45).

For some animals, particularly young dogs, damage of thevessel wall and stenosis are not sufficient to induce thromboticcyclic flow variations. In these cases, additional activation ofplatelets by infusion of epinephrine (protocol 3) is required,leading to the formation of measurable thrombi. In protocol 4,thrombus formation is induced by subcritical stenosis withoutprior clamping of the artery and infusion of platelet activatingfactor (PAF), according to the model described by Apprill et al.(46). In addition to these protocols, coronary spasms induced byreleased platelet components can influence CBF. Therefore, thismodel includes the main pathological factors of unstable anginapectoris.

2.1.1.2. CoronaryThrombosis Induced byElectrical Stimulation(Protocol 5)

In this protocol, coronary thrombosis is induced by delivery oflow amperage electrical current to the intimal surface of theartery, as described by Romson et al. (47). In contrast to thestenosis protocols, an occluding thrombosis is formed graduallywithout embolism after some hours. As a consequence of the timecourse, thrombi formed are of mixed type and contain more fibrinthan platelet thrombi formed by critical stenosis.


2.1.2. Procedure

2.1.2.1. Protocol 1:Critical Stenosis

Dogs of either sex weighing 15–40 kg and at least 8 monthsof age are anesthetized with pentobarbital sodium (bolus of30–40 mg/kg and then continuous infusion of approximately0.1 mg/kg/min); respiration is maintained through a trachealtube using a positive pressure respirator. The heart is exposedthrough a left thoracotomy at the fourth or fifth intercostal space;the pericard is opened and the left circumflex coronary artery(LCX) is exposed. An electromagnetic or Doppler flow probe isplaced on the proximal part of the LCX to measure CBF. Distalto the flow probe, the vessel is squeezed with a 2-mm hemostaticclamp for 5 s. A small cylindrical plastic constrictor 2–4 mm inlength and with an internal diameter of 1.2–1.8 mm (dependingon the size of the LCX) is then placed around the artery at the siteof the damage. Usually, the constrictor has to be changed severaltimes (2–5 times) until the appropriate narrowing of the vessel isachieved and cyclic flow variations are observed. In the event ofocclusion of the artery without spontaneous embolization of theformed thrombus, reflow is induced by shortly lifting the vesselwith a thread placed beneath the stenotic site.

Only dogs with regularly spaced CFRs of similar inten-sity within a pre-treatment phase of 60 min are used inthese experiments. The test substance is administered by i.v.bolus injection or continuous infusion, or by intraduodenalapplication. CFRs are registered for 2–4×60 min and com-pared to pre-treatment values. Prior to testing, preparationsfor additional hemodynamic measurements are performed (seebelow).

2.1.2.2. Protocol 2:Subcritical Stenosis

Male castrated pigs (German landrace weighing 20–40 kg)are anesthetized with ketamine (2 mg/kg i.m.), metomidate(10 mg/kg i.p.), and xylazine (1–2 mg/kg i.m.). In order tomaintain the stage of surgical anesthesia, animals receive a contin-uous i.v. infusion of 0.1–0.2 mg/kg/min pentobarbital sodium.Respiration is maintained through a tracheal tube using a posi-tive pressure respirator. The heart is exposed through a left tho-racotomy at the fourth and fifth intercostal space; the pericard isopened and the LAD is exposed. An electromagnetic or Dopplerflow probe is placed on the proximal part of the LAD to mea-sure CBF. Distal to the flow probe, the vessel is squeezed with a1-mm hemostatic clamp for 5 s. A small cylindrical plastic con-strictor 2 mm in length is then placed around the artery at thesite of damage. Usually, the constrictor has to be changed severaltimes until the appropriate narrowing of the vessel is achievedthat produces CFRs. CFRs are similar to those in dogs; pigs,however, show a reactive hyperemic response. If embolization

42 Mousa

does not occur spontaneously, the formed thrombus is releasedby reducing blood flow by shortly lifting the vessel with forceps.

Only pigs with regularly spaced CFRs of similar intensitywithin a pre-treatment phase of 60 min are used for the exper-iments. The test substance is administered by i.v. bolus injectionor continuous infusion, or by intraduodenal application. CFRs areregistered for 2×60 min and compared to pre-treatment values.

2.1.2.3. Protocol 3:Stenosis PlusEpinephrine Infusion

If protocol 1 does not lead to CFRs, additional epinephrine(0.2 μg/kg/min) is infused into a peripheral vein for 2×60 min(60 min before and 60 min after drug administration). CFRs arerecorded, and the 60-min post-drug phase is compared to the60-min pre-drug phase.

2.1.2.4. Protocol 4:Stenosis Plus PAFInfusion

The LCX is stenosed without prior mechanical wall injury. Thispreparation does not lead to thrombus formation (subcriticalstenosis). For the induction of CFRs, PAF (C16-PAF, Bachem)(0.2 nmol/kg/min) is infused into one cannulated lateral branchof the coronary artery. After 30 min, PAF infusion is terminatedand blood flow returns to a normal, continuous course. Thirtyminutes later, the test substance is administered concomitantlywith the initiation of a second PAF infusion for 30 min. CFRs arerecorded and the drug treatment/second PAF phase is comparedto the pre-drug/first PAF phase.

2.1.2.5. Protocol 5:Electrical Stimulation

The LCX is punctured distal to the flow probe with a chrome–vanadium–steel electrode (3 mm in length, 1 mm diameter). Theelectrode (anode) is placed in the vessel in contact with the inti-mal lining and connected over a teflon-coated wire to a 9-volt(V) battery, a potentiometer, and an amperemeter. A disc elec-trode (cathode) is secured to a subcutaneous thoracal muscle layerto complete the electrical circuit. The intima is stimulated with150 μA for 6 hours (h). During this time, an occluding throm-bosis is gradually formed.

The test substance, or vehicle as a control, is administeredeither at the start of the electrical stimulation or 30 min after thestart. The time until thrombotic occlusion of the vessel occursand the thrombus size (wet weight measured immediately afterremoval at the end of the experiment) are determined. Prior totesting, preparations for additional hemodynamic measurementsare performed (see below).

For all protocols the following preparations and measure-ments are performed:

1. To measure peripheral arterial blood pressure (BP) [mmHg], the right femoral artery is cannulated and connectedto a Statham pressure transducer.


2. Left ventricular pressure (LVP) [mm Hg] is determined byinserting a micro tip catheter via the carotid artery retro-gradely.

3. Left ventricular end-diastolic pressure (LVEDP) [mm Hg]is evaluated through sensitive amplification of the LVP.

4. Contractility (LV dp/dt max) [mm Hg/s] is determinedfrom the initial slope of the LVP curve.

5. Heart rate [min–1] is determined from the pulsatile bloodpressure curve.

6. The ECG is recorded in lead II.7. Arterial pH and concentrations of blood gases are main-

tained at physiological levels by adjusting respiration andinfusion of sodium bicarbonate.

8. Blood hematocrit values (37–40%) and number of erythro-cytes are kept constant by infusion of oxypolygelatine indogs and electrolyte solution in pigs.

9. Body temperature is monitored with a rectal thermistorprobe and kept constant by placing the animals on a heatedmetal pad with automatic temperature regulation.

10. Template buccal mucosal bleeding time is carried out usingthe Simplate device.

11. At the end of the test, animals are sacrificed by an overdoseof pentobarbital sodium.

For detailed applications of the Folts model, see Folts (44),Folts and Rowe (42, 43), and Folts et al. (19, 24).

2.1.3. Evaluation For all protocols, the mean maximal reduction of blood pressure(systolic/diastolic) [mm Hg] is determined.

2.1.3.1. Protocols 1–4 The following parameters are determined to quantify stenosis-induced coronary thrombosis:

1. Frequency of CFRs = cycle number per unit time2. Magnitude of CFRs = cycle area [mm2] (total area of all

CFRs per unit time, measured by planimetry)3. Percent change in cycle number and cycle area after drug

treatment is calculated relative to pre-treatment controls.4. Statistical significance is assessed by the paired Student’s

t-test.

2.1.3.2. Protocol 5 The following parameters are determined to quantify electricallyinduced coronary thrombosis:

1. Occlusion time [min] = time to zero blood flow.2. Thrombus size [mg] = wet weight of the thrombus imme-

diately after removal.

44 Mousa

3. Percent change in mean values for occlusion time andthrombus size in drug-treated groups is calculated relativeto the control group.

4. Statistical significance is assessed by the non-paired Student’st-test.


The stenosis (Folts) and electrical (Romson/Lucchesi) mod-els of coronary thrombosis are widely used to study the roleof mediators in the thrombotic process and the effect of newantithrombotic drugs. Bush and Patrick (28) provide an excel-lent review of the role of the endothelium in arterial thrombo-sis and the use of the Folts model to determine the effects ofthrombosis inhibitors and mediators, i.e., thromboxane, prosta-cyclin, cyclooxygenase, serotonin, nitric oxide (NO) donors, andother vasodilators. The effect of an NO donor could be reversedby the NO scavenger oxyhemoglobin, which indicated that NOindeed mediates antithrombotic drug action (48). These coro-nary thrombosis models have recently been used to elucidate themechanisms of action of several antithrombotic drugs, includingthe oral GPIIb/IIIa antagonist DMP 728 (49); the low molecularweight heparin (LMWH) enoxaparin (50), which, in contrast tounfractionated heparin, inhibited CFRs; the thrombin inhibitorPEG–hirudin (51); melagatran (52), an anti-P-selectin antibody(53); and activated protein C (54).

The clinical relevance of the Folts model has been questionedbecause the model is very sensitive to antithrombotic compounds.However, in this model, lack of a reversal by epinephrine or anincrease in degree of stenosis is able to differentiate any new drugfrom aspirin. Electrical coronary thrombosis is less sensitive (i.e.,aspirin has no effect) and higher doses of some drugs are required.However, in principle, most drugs act in both models, if at all.

2.1.5. Modificationsof the Method

Romson et al. (55) described a simple technique for the inductionof coronary artery thrombosis in the conscious dog by delivery oflow amperage electric current to the intimal surface of the artery.

Benedict et al. (56) modified the electrical stimulation ofthrombosis model by using two Doppler flow probes proximaland distal to the needle electrode in order to measure changesin blood flow velocity. The electrical current was stopped when a50% increase in flow velocity was reached, at which point throm-bosis occurred spontaneously. Using this model, the investiga-tors demonstrated the importance of serotonin by measuringincreased coronary sinus serotonin levels just prior to occlusion.

Warltier et al. (57) described a canine model of thrombin-induced coronary artery thrombosis to analyze the effects ofintracoronary SK on regional myocardial blood flow, contrac-tile function, and infarct size. Al-Wathiqui et al. (36) described


the induction of CFRs in the coronary, carotid, and femoralarteries of conscious chronically instrumented dogs. The Foltsthrombosis model has also been applied to carotid arteriesin monkeys. Coller et al. (58) induced CFRs in the carotidarteries of anesthetized cynomologus monkeys and demon-strated that they were abolished by the GPIIb/IIIa antibodyabciximab.

2.2. Stenosis- andMechanicalInjury-InducedArterial and VenousThromboses(Harbauer Model)


Harbauer (59) first described a venous model of thrombosisinduced by mechanical injury and stenosis of the jugular vein.In a modification of the technique, both arterial and venousthromboses are produced in rabbits by stenosis of the carotidartery and the jugular vein with simultaneous mechanical dam-age of the endothelium. This results in the activation of plateletsand the coagulation system and leads to changes in the blood-stream pattern. As a consequence, occluding thrombi are formedand detected by blood flow measurements. The dominant roleof platelets in this model is evidenced by the inhibitory effect ofan antiplatelet serum in both types of vessels (60). The modifiedHarbauer model is used to evaluate the antithrombotic activity ofcompounds in an in vivo model of arterial and venous thrombosesin which thrombus formation is highly dependent on platelet acti-vation.

2.2.2. Procedure Male Chinchilla rabbits weighing 3–4 kg receive test compoundor vehicle as a control by oral, i.v., or i.p. administration. Thefirst ligature (vein; for preparation, see below) is performed at theend of the absorption period (i.p., approximately 30 min; p.o.,approximately 60 min; i.v., variable).

Sixty-five minutes before stenosis, the animals are sedated byi.m. injection of 8 mg/kg xylazine (Rompun R©) and anesthetizedby i.v. injection of 30–40 mg/kg pentobarbital sodium 5 minlater. During the course of the test, anesthesia is maintained bycontinuous infusion of pentobarbital sodium (30–40 mg/kg/h)into one femoral vein.

A Statham pressure transducer is placed into the right femoralartery for continuous measurement of blood pressure. Sponta-neous respiration is maintained through a tracheal tube. Onejugular vein and one carotid artery are exposed on opposite sides.Small branches of the vein are clamped to avoid blood flowaround the vessel occlusion.

46 Mousa

Electromagnetic or Doppler flow probes are placed on thevein (directly central to the vein branching) and on the artery(as centered as possible). Blood flow [ml/min] is measuredcontinuously.

After blood flow reaches a steady state (approximately15–30 min), a metal rod with a diameter of 1.3 mm is placedon the jugular vein (2 cm central to the vein branching) and a lig-ature is tightened. After 1 min, the rod is removed from the liga-ture. Immediately thereafter (approximately 1.5 min), the carotidartery is damaged by briefly squeezing it with forceps. A smallplastic constricting cylinder (2 mm in length and 1.2 mm in diam-eter) is placed around the site of endothelial damage.

Template bleeding time is measured at various time inter-vals before and after drug treatment (depending on the route ofadministration) in the shaved inner ear using a Simplate R© device.Care is taken to select parts of the skin without large vessels.

2.2.3. Evaluation 1. Percent thrombus formation (thrombosis incidence) isdetermined as the number of occluded vessels (bloodflow=0).

2. Percent inhibition of thrombosis is calculated in each dosagegroup relative to the respective vehicle controls.

3. Thrombosis incidence in the vehicle controls is set as 100%.4. Statistical significance is assessed by means of the Fisher-

exact-test.5. If initial values for blood flow do not significantly differ in

the dosage and control groups, the area below the bloodflow curve is measured by planimetry, and the mean valueof each dosage group is compared to the control using theunpaired Student’s t-test.

6. Mean occlusion time [min] in the dosage and control groupsare calculated and compared using the Students’s t-test.

7. The maximal change in systolic and diastolic blood pressureduring the time period of stenosis as compared to the ini-tial values before drug administration is determined. Thereis no standardized assessment score. For example, a reduc-tion of systolic blood pressure by 30 mmHg and diastolicblood pressure by 20 mmHg is generally accepted as a strongreduction in blood pressure.


Two main factors of arterial thrombosis in human are essentialcomponents of this model: high-grade stenosis and vessel walldamage. In the absence of either, no thrombus is formed. Theocclusive thrombus is formed fast and in a highly reproduciblemanner. In both vessels, thrombus formation is dependent onplatelet function, as shown by the effects of antiplatelet serum.


Thus, jugular vein thrombosis in this model differs from stasis-induced deep vein thrombosis with prominent fibrin formation.On the other hand, occlusive thrombi are more stable than thepure platelet thrombi in the Folts model (see Section 2.1), ascarotid blood flow cannot be restored by shaking the constric-tor. The following antithrombotic drugs have been shown tobe effective in this model: (1) antiplatelet drugs such as ticlopi-dine, prostacyclin/iloprost, NO donors (SNP, molsidomine), butnot aspirin or thromboxane synthetase inhibitors; (2) anticoag-ulants such as hirudin, high-dose heparin, and warfarin; and (3)SK/t-PA (60, 61, and unpublished data). Drugs that simply lowerblood pressure, such as hydralazine, clonidine, and prazosin haveno effect on thrombus formation in this model.

2.2.5. Modifications ofthe Method

Bevilacqua et al. (61) applied this model to the rabbit carotidarteries and compared one artery before drug treatment withthe contralateral artery after drug treatment. Heparin, the syn-thetic thrombin inhibitor FPRCH2Cl, iloprost, and t-PA, but notaspirin, inhibited carotid occlusion in this model.

Spokas and Wun (62) induced venous thrombosis in the venacava of rabbits by vascular damage and stasis. Vascular wall dam-age was achieved by crushing the vessel with hemostat clamps. Asegment of the vena cava was looped with two ligatures 2.5-cmapart, and then 2 h after ligation, the isolated venous sac was dis-sected and the clot was removed for determination of dry weight.

Lyle et al. (63), in pursuit of an animal model that mimickedthrombotic re-occlusion and restenosis after successful coronaryangioplasty in human, developed a model of angioplasty-inducedinjury in atherosclerotic rabbit femoral arteries. Acute 111indium-labeled platelet deposition and thrombosis were assessed 4 h afterballoon injury in arteries subjected to prior endothelial dam-age (by air desiccation) and cholesterol supplementation (onemonth). The effects of inhibitors of FXa or platelet adhesion, hep-arin, and aspirin on platelet deposition were studied.

Meng (64), Meng and Seuter (65), and Seuter et al. (66)described a method to induce arterial thrombosis in rats by chill-ing of the carotid artery (thrombosis induced by super cool-ing). Rats were anesthetized, and then the left carotid artery wasexposed and occluded proximal by means of a small clamp. Theartery was placed for 2 min into a metal groove that was cooled to–15◦C. The vessel was then compressed using a weight of 200 g.In addition, a silver clip was fixed to the vessel distal to the injuredarea to produce disturbed and slow blood flow. After 4 min, theproximal clamp was removed and blood flow was reestablished inthe injured artery. A similar model in the rabbit has also beendeveloped, with slightly different conditions (chilling tempera-ture of –12◦C for 5 min, and a compression weight of 500 g).The wound is closed and the animal is allowed to recover from

48 Mousa

anesthesia. Antithrombotic compounds are administered in var-ious doses at different time intervals before surgery. After 4 h,the animals receive heparin and are re-anesthetized. The lesionedcarotid artery is removed and thrombus wet weight is immedi-ately measured.

2.3. ElectricallyInduced Thrombosis

A novel technique for inducing arterial thrombosis was intro-duced by Salazar (67) in which anodal current was delivered tothe intravascular lumen of a coronary artery in the dog via astainless steel electrode. The electrode was positioned under flu-oroscopic control, which somewhat complicated the procedure.Subsequently, Romson et al. (55) modified the procedure suchthat the electrode was placed directly into the coronary artery ofan open-chest, anesthetized dog. This technique then allows oneto produce a thrombus in the anesthetized animal or to close thechest after inserting the electrode and allow the animal to recover,after which thrombosis can be elicited later in the conscious ani-mal. The advantage of this modification is that it allows inductionof thrombus formation without the need for fluoroscopy.

The stimulation electrode is constructed from a 25- or 26-gauge stainless steel hypodermic needle tip, which is attached to a30-gauge teflon-insulated silver-coated copper wire. Anodal cur-rent is delivered to the electrode via either a 9-V nickel–cadmiumbattery with the anode connected in series to a 250,000-ohmpotentiometer or with a Grass stimulator connected to a Grassconstant current unit and a stimulus isolation unit. The cathodein both cases is placed into a subcutaneous site completing the cir-cuit. The anodal current can be adjusted to deliver 50–200 μA.Anodal stimulation results in focal endothelial disruption, whichin turn induces platelet adhesion and aggregation at the damagedsite. This process is then followed by further platelet aggrega-tion and consolidation, with the growing thrombus entrappingred blood cells.

A modification of the method of Romson et al. (55) involvesplacement onto the coronary artery of an external, adjustableoccluder (68) to produce a fixed stenosis on the coronary artery.A flow probe to record CBF is placed on the proximal portion ofthe artery followed by the stimulation electrode, with the clampbeing placed most distally (Fig. 2.2). The degree of stenosis canthen be controlled by adjusting the clamp. The resulting steno-sis is produced in an effort to mimic the human pathophysiologyof atherosclerotic coronary artery disease, whereby thrombolytictherapy restores CBF through a coronary artery with residual nar-rowing due to atherosclerotic plaque formation.

Another modification of the electrical stimulation model thatmerits discussion is described by Benedict et al. (56). They dis-continued anodal current when mean distal coronary flow velocity(measured with Doppler flow meter) increased by approximately


Fig. 2.2. Model of coronary artery thrombosis in the dog. Electrical injury to the intimalsurface of the artery leads to occlusive thrombus formation. The thrombus is formed inthe presence of a flow-limiting stenosis induced by a Goldblatt clamp. Upon spontaneousocclusion, heparin is administered and the clot is aged for 1 h before initiating the t-PAinfusion.

50%, reflecting disruption of normal axial flow by the growingthrombus. Occlusive thrombosis occurred within 1 h after stop-ping the current (2 h after starting the current). In these studies,coronary sinus plasma levels of serotonin, an index of intravas-cular platelet aggregation, were increased approximately 20-foldjust before occlusive thrombus formation. The results of thesestudies agree with others in showing that either proximal flowvelocity or electromagnetically measured CBF declines triviallyover the majority of the time period in which the thrombus isgrowing. The largest declines in (volume) flow occur over a smalland terminal fraction of the period between initial vessel pertur-bation and final occlusion. During that interval, coronary lumenalarea decreases rapidly and to a critical degree, as platelets accrueat the growing thrombus. The studies by Benedict et al. (56)demonstrate that this final phase of thrombosis can occur inde-pendently of electrical stimulation. This variation of the modelmay be attractive to those who wish to produce occlusive throm-bosis without continued electrical stimulation.

50 Mousa

Regardless of whether electrical stimulation is continued untilocclusive thrombosis, there is another component to this modelthat has upside and downside potential, namely, the opportunityfor coronary vasoconstriction to occur. Although the incidence ofPrinzmetal’s angina is low, it is widely suspected that vasospasmsuperimposes on a primarily thrombotic event in unstable anginaand myocardial infarction. In studies by Van der Giessen et al.(69), nifedipine was reported to increase the extent of CBF afterplasmin-induced thrombolysis in a porcine model of electricallyinduced coronary thrombosis. In their model, the anodal stimu-lation was applied circumferentially to the exterior surface of theLAD, and an external constrictor was not used.

Depending on the hypothesis being tested, the experimentercan leave intact or minimize this potential through the use ordisuse of an external constrictor. As in the Folts and Gold coro-nary thrombo(ly)sis models, blood pressure must be taken intoaccount or maintained within acceptable limits, since, in the pres-ence of a critical stenosis, auto-regulation no longer exists. Underthese conditions CBF is highly dependent on driving pressure(arterial pressure).

Numerous experimental studies evaluating anticoagulants,antithrombotic, and/or thrombolytic drugs have been performedusing this model. In the initial report by Romson et al. (55), thecyclooxygenase inhibitor ibuprofen was evaluated. Comparisonof myocardial infarct size, thrombus weight, arrhythmia devel-opment, and scanning electron microscopy of drug-treated andcontrol animals indicated that ibuprofen protected the consciousdog against the deleterious effects of coronary artery thrombo-sis. Subsequent studies in the same model and laboratory eval-uated the antithrombotic potential of various TXA2 synthetaseinhibitors, such as U 63557A, CGS 13080, OKY 1581, anddazoxiben. When the TXA2 synthetase inhibitors were admin-istered before induction of the current, OKY 1581 (70) andCGS 13080 (71) reduced the incidence of coronary thrombosis,whereas U 63557A (72) and dazoxiben (73) were ineffective andpartially effective, respectively. The differences in efficacy notedamong the TXA2 synthetase inhibitors were ascribed to differ-ences in potency and duration of action.

Other investigators have used this model to study the preven-tion of original coronary thrombosis in the dog. Fitzgerald et al.(74) studied the TXA2 synthetase inhibitor U 63557A alone orin combination with L-636,499, an endoperoxide/thromboxanereceptor antagonist. U 63557A alone did not prevent coro-nary thrombosis when administered before current application,whereas the combination of U 63557A and L-636,499 was highlyeffective. These data suggest that prostaglandin endoperoxidesmay modulate the effects of TXA2 synthetase inhibitors and thatthis response may be blocked by concurrent administration of


an endoperoxide/thromboxane receptor antagonist. The murinemonoclonal antibody to platelet GPIIb/IIIa (7E3) was studiedin this model for its ability to prevent thrombus formation. At adose of 0.8 mg/kg i.v., the 7E3 monoclonal antibody completelyprevented original thrombus formation (75).

In addition to the evaluation of antithrombotic (i.e.,antiplatelet) agents, the electrical injury model is useful for study-ing anticoagulant FXa inhibitors, such as YM 60628 (86), andthrombolytic drugs. When evaluating thrombolytic agents, thethrombus is allowed to form without drug intervention and thenaged for various periods. Schumacher et al. (68) demonstratedthat intracoronary SK was an effective thrombolytic drug in thismodel; the thrombolytic effectiveness being augmented by theconcurrent administration of heparin and prostacyclin, or by aTXA2 synthetase inhibitor (68). In other studies reported byShebuski et al. (76), the TXA2 receptor antagonist BM 13.177hastened t-PA-induced thrombolysis and prevented acute throm-botic re-occlusion. Van der Giessen et al. (77) subsequentlydemonstrated that BM 13.177 prevented original thrombus for-mation in 75% of pigs undergoing electrical stimulation; aspirinwas ineffective in this porcine model. These and other studiesunderscore the potential for adjunctive therapy to hasten throm-bolysis and/or prevent re-occlusion, both contributing to greatersalvage of ischemic myocardium.

Like the copper coil model, the electrical stimulation modelhas been used to produce experimental myocardial infarction.Patterson et al. (78) have used this technique to produce coro-nary thrombosis in the LCX (which supplies blood flow to theposterior LV wall in dogs) in dogs with a previous anteriorwall infarct to mimic sudden cardiac death that occurs in peo-ple during a second (recurrent) myocardial infarction or ischemicevent.

This model has also been modified to demonstrate the effi-cacy of adjuncts to thrombolytic therapy (79–82). In this case,the thrombus is allowed to extend until it completely occludes thevessel. Usually, the thrombus is allowed to stabilize, or “age,” tomimic the clinical setting in which a time lag exists between thethrombotic event and the pharmacological intervention. At theend of the stabilization period, thrombolytic agents such as t-PAor SK are administered in conjunction with the novel antithrom-botic agent to lyse the thrombus and maintain vessel patency. Theincidence and times of reperfusion and re-occlusion are the majorendpoints. These studies have established that recombinant tickanticoagulant peptide (rTAP), a potent and selective FXa inhibitorderived from the soft tick (83), promotes rapid and prolongedreperfusion at doses that produce relatively minor elevations inPT, aPTT, and template bleeding time.

52 Mousa


The use of an electrical current to induce thrombosis in hamsterand dog was described in the early 1950s by Lutz et al. (84) andSawyer et al. (85, 86). In general, two different approaches aretaken in this model. One method produces electrical damage bymeans of two externally applied hook-like electrodes (87, 88).The other method uses a needle electrode that is advancedthrough the walls of the blood vessel and positioned in the lumen;a second electrode is placed at a subcutaneous site to complete thecircuit (55, 56, 67).

2.3.2. Procedure Anaesthetized rats weighing 200–300 g are intubated and afemoral artery is cannulated for administration of test com-pound(s). One carotid artery is isolated from the surroundingtissue over a distance of 10–15 mm.

A pair of rigid stainless steel wire hook-like electrodes with aworking distance of 4 mm is positioned on the artery by meansof a rack and pinion gear manipulator. The artery is raised slightlyaway from the surrounding tissue. Isolation of the electrodes isachieved by the insertion of a small piece of parafilm under theartery. Blood flow is measured with an ultrasonic Doppler flowmeter (Transonic, Ithaca NY, USA); the flow probe (1RB) isplaced proximal to the damaged area.

Thrombus formation is induced in the carotid arteries by theapplication of an electrical current (350 V, DC, 2 mA) deliv-ered by an electrical stimulator (Stoelting Co, Chicago, Cat.No 58040) for 5 min to the exterior surface of the artery.

2.3.3. Evaluation 1. Blood flow before and after induction of thrombus for60 min.

2. Time to occlusion [min] = the time between onset of theelectrical current and the time at which blood flow decreasesto less than 0.3 ml/min.

3. Patency of the blood vessel over 30 min.


Thrombi formed by electrical induction are composed of denselypacked platelets, with some red cells. Moreover, electrical injurycauses extensive damage to intimal and sub-intimal layers. Theendothelium is completely destroyed, and the damage extendsto sub-endothelial structures, including smooth muscle cells.This deep damage could reduce sensitivity in terms of dis-criminating between drugs on the basis of their antithromboticactivity. However, Philp et al. (88) showed that unfractionatedheparin completely blocks thrombus formation, whereas otherantiplatelet agents exhibit differential antithrombotic actions. Theinvestigators concluded that this relatively simple model of arte-rial thrombosis might prove to be a useful screening test for drugswith antithrombotic potential.



In a modification of this model by Salazar (67), a stainless steelelectrode is inserted into a coronary artery in the dog to deliveranodal current to the intravascular lumen. The electrode is posi-tioned under fluoroscopic control, complicating the proceduresomewhat. Romson et al. (55) described a further modification inwhich the electrode was placed directly into the coronary arteryof open-chest anaesthetized dogs.

Rote et al. (89, 90) applied the carotid thrombosis model todogs. A calibrated electromagnetic flow meter was placed on eachcommon carotid artery proximal to the point of insertion of anintravascular electrode and a mechanical constrictor. The externalconstrictor was adjusted with a screw until the pulsatile flow pat-tern was decreased by 25% without alteration in mean blood flow.Electrolytic injury to the intimal surface was accomplished withan intravascular electrode composed of a teflon-insulated silver-coated copper wire connected to the positive pole of a 9-V nickel–cadmium battery in series with a 250,000-ohm variable resistor.The cathode was connected to a subcutaneous site. Injury was ini-tiated in the right carotid artery by application of a 150-μA con-tinuous pulse anodal direct current to the intimal surface of thevessel for a maximum duration of 3 h, or for 30 min beyond thetime of complete vessel occlusion, as determined by blood flowrecordings. Upon completion of the study on the right carotid,the procedure was repeated on the left carotid artery after admin-istration of test drug.

Benedict et al. (56) introduced a procedure in which anodalcurrent was discontinued when mean distal coronary flow velocityincreased by approximately 50%, reflecting disruption of normalflow by the growing thrombus. An occlusive thrombosis occurredwithin 1 h after cessation of the electrical current. In this model,the final phase of thrombosis occurred independently of electricalinjury.

A ferret model of acute arterial thrombosis was developedby Schumacher et al. (91). A 10-min anodal electrical current of1 mA was delivered to the external surface of the carotid arterywhile measuring carotid blood flow. This produced an occlusivethrombus in all vehicle treated ferrets within 41±3 min with anaverage weight of 8±1 mg. Thrombus weight was reduced byaspirin or a thromboxane receptor antagonist.

Guarini (92) reported the formation of a completely occlu-sive thrombus in the common carotid artery of rats by applyingan electrical current to the arterial wall (2 mA for 5 min) whilesimultaneously constricting the artery with a hemostatic clampplaced immediately downstream from the electrodes.

2.4. Ferric Chloride(FeCl3)-InducedThrombosis

The administration of a variety of chemicals either systemi-cally or locally can result in damage to the endothelium with

54 Mousa

subsequent generation of a thrombus. Such compounds includeferric/ferrous chloride, fluorescein-labeled dextran and RoseBengal.

In models employing ferric (ferrous) chloride (93), thecarotid artery of rats is isolated. A flow probe is placed proximalto the intended site of lesion and a 3-mm disc of filter paper whichhas been soaked in ferric/ferrous chloride (35-50%) is placed onthe artery. The application of ferric (ferrous) chloride results intransmural vascular injury leading to the formation of occlusivethrombi. This injury is believed to be a result of lipid peroxidationcatalyzed by the ferric (ferrous) chloride. Thrombus formation,measured as a decrease in blood flow through the vessel, typicallyoccurs within 30 min. Microscopic analysis of the thrombi hasshown them to be predominantly platelet-rich clots. This modelhas been used to study the antithrombotic effects of direct throm-bin inhibitors (94–96) and heparins.

Endothelial damage can also be induced by fluorescein-or fluorescein isothiocyanate (FITC)-conjugated compounds. Amodel has been described in which FITC–dextran is admin-istered intravenously to mice. Thrombus formation is inducedupon exposure of the microvessels of the ear to the light ofa mercury lamp (excitation wavelength of 450–490 nm) (97).The endothelial damage induced in this model is believed tobe a result of the generation of singlet molecular oxygen pro-duced by energy transfer from the excited dye (98). Thrombusformation is measured using intravital fluorescence microscopy.This detection technique allows for a number of endpoints tobe quantitated, including changes in luminal diameter due tothrombus formation, blood flow measurements, and extravasa-tion of the FITC–dextran. This model offers the advantages ofnot requiring surgical manipulations, which can cause hemo-dynamic or inflammatory changes, allowing for repeated anal-ysis of the same vessel segments over time, and being appli-cable to the study of both arteriolar and venular thromboses.The administration of Rose Bengal has been used similarly(99).


A variety of chemical agents have been used to induce thrombosisin animals. The use of topical FeCl3 as a thrombogenic stimulus inveins was described by Reimann-Hunziger (100). Kurz et al. (93)demonstrated that the thrombus produced with this method inthe carotid arteries of rats is composed of platelets and red bloodcells enmeshed in a fibrin network. This simple and reproducibletest has been used for the evaluation of antithrombotic (101) andpro-fibrinolytic test compounds (102).

2.4.2. Procedure Rats weighing 250–300 g are anaesthetized with Inactin(100 mg/kg) and a polyethylene catheter (PE-205) is inserted


into the trachea via a tracheotomy to facilitate breathing.Catheters are also placed in the femoral artery for blood sam-pling and measurement of arterial blood pressure and in the jugu-lar vein for administration of test compounds. The right carotidartery is isolated and an ultrasonic Doppler flowprobe (probe1RB, Transonic, Ithaca NY, USA) is placed on the vessel to mea-sure blood flow. A small piece of parafilm “M” (American CanCo., Greenwich, CT, USA) is placed under the vessel to isolate itfrom surrounding tissue throughout the experiment.

The test compound is administered by gavage or as an i.v.injection at a defined time prior to initiation of thrombus for-mation. Thrombus formation is induced by the application of apiece of filter paper (2 mm×5 mm) saturated with 25% FeCl3 tothe carotid artery. The paper is allowed to remain on the vesselfor 10 min and is then removed. Parameters (see below) are mon-itored for 60 min after the induction of thrombosis, after whichthe thrombus is removed and weighed.

2.4.3. Evaluation 1. Blood flow before and after induction of thrombus for60 min

2. Time to occlusion [min]: the time between FeCl3 applica-tion and the point at which blood flow decreases to less than0.3 ml/min

3. Thrombus weight after blotting the thrombus on filter paper

2.5.Thrombin-InducedClot Formation inRabbit Femoral orCanine CoronaryArtery

Localized thrombosis can also be produced in rabbit peripheralblood vessels such as the femoral artery by injection of thrombin,calcium chloride and fresh blood via a side branch (103).

Either femoral artery is isolated distal to the inguinal liga-ment and traumatized distally from the lateral circumflex arteryby rubbing the artery with the jaws of forceps. An electromag-netic flow probe is placed distal to the lateral circumflex arteryto monitor femoral artery blood flow (Fig. 2.3). The superficialepigastric artery is cannulated for induction of the thrombus andsubsequent infusion of thrombolytic agents. Localized thrombidistal to the lateral circumflex artery with snares approximately1-cm apart are induced by the sequential injection of thrombin,CaCl2 (1.25 mmol), and a volume of blood sufficient to distendthe artery. After 30 min, the snares are released and femoral arteryblood flow is monitored for 30 mm to confirm total obstructionof flow by the thrombus.

These models are not appropriate for evaluating drugs fortheir ability to inhibit original thrombosis. However, the modelis particularly appropriate for evaluating thrombolytic agents andadjunctive therapies for their ability to hasten and/or enhancelysis or prevent acute re-occlusion after discontinuing administra-tion of a thrombolytic agent.

56 Mousa

Fig. 2.3. Rabbit model of femoral arterial thrombosis. A clot is introduced into an isolated segment of femoral artery byinjection of thrombin, CaCl2, and whole blood. After aging for 1 h, t-PA is infused. Reperfusion is assessed by restorationof blood flow.


A canine model of thrombin-induced clot formation was devel-oped by Gold et al. (104) in which localized coronary throm-bosis was produced in the LAD. This is a variation of the tech-nique described by Collen et al. (105) who used radioactivefibrinogen to monitor the occurrence and extent of thrombo-lysis of rabbit jugular vein clots. The vessel was intentionallyde-endothelialized by external compression with blunt forceps.Snare occluders were then placed proximal and distal to the dam-aged site, and thrombin (10 U) was injected into the isolatedLAD segment in a small volume via a previously isolated sidebranch. Autologous blood (0.3–0.4 ml) mixed with calcium chlo-ride (0.05 M) was also injected into the isolated LAD segment,producing a stasis-type red clot superimposed on an injured bloodvessel. The snares were released 2–5 min later and total occlusionwas confirmed by selective coronary angiography. This model ofcoronary artery thrombosis relies on the conversion of fibrinogento fibrin by thrombin. The fibrin-rich thrombus contains platelets,but at no greater concentration than in a similar volume of wholeblood. Once the thrombus is formed, it is allowed to age for1–2 h, after which a thrombolytic agent can be administered tolyse the thrombus and restore blood flow.


2.5.2. Procedure In the initial study described by Gold et al. (104), recombinantt-PA was characterized for its ability to lyse 2-h-old thrombi.Tissue plasminogen activator was infused at doses of 4.3, 10, and25 μg/kg/min i.v. and resulted in reperfusion times of 40, 31,and 13 min, respectively. Thus, in this model of canine coronarythrombosis, t-PA exhibited dose-dependent coronary thromboly-sis. It is also possible to study the effects of different doses of t-PAon parameters of systemic fibrinolytic activation, such as fibrino-gen, plasminogen, and a2-antiplasmin, as well as to assess myocar-dial infarct size. For example, Kopia et al. (106) demonstratedthat SK elicited dose-dependent thrombolysis in this model.

Subsequently, Gold et al. (107, 108) modified the modelto study not only reperfusion, but also acute re-occlusion.Clinically, re-occlusion is a persistent problem after effectivecoronary thrombolysis, which is reported to occur in 15–45%of patients (109). Thus, an animal model of coronary reper-fusion and re-occlusion would be important from the stand-point of evaluating adjunctive therapies to t-PA to hasten and/orincrease the response rate to thrombolysis as well as prevent acutere-occlusion.

The model of thrombin-induced clot formation in the caninecoronary artery was modified such that a controlled high-gradestenosis was produced with an external constrictor. Blood flowwas monitored with an electromagnetic flow probe. In this modelof clot formation with superimposed stenosis, reperfusion inresponse to t-PA occurs with subsequent re-occlusion. The mon-oclonal antibody against the human GPIIb/IIIa receptor devel-oped by Coller et al. (110) and tested in combination with t-PA inthe canine thrombosis model hastened t-PA-induced thromboly-sis and prevented acute re-occlusion (111). These actions in vivowere accompanied by abolition of ADP-induced platelet aggrega-tion and markedly prolonged bleeding time.

2.6. Laser-InducedThrombosis

The physiologic responses to injury in the arterial and venoussystems vary in part due to differences in blood flow conditions,leading to different clot compositions. This model of arterialthrombosis is based on the development of a platelet-rich throm-bus following laser-mediated thermal injury to the vascular wall.This model was first described by Weichert and Breddin (112).In this model, an intestinal loop of an anesthetized rat is exposedthrough a hypogastric incision and spread on a microscope stagewhile being continuously irrigated with sterile physiologic saline.Vascular lesions are induced on small mesenteric arterioles withan argon laser beam (50 mW at microscope, 150-ms duration)directed through the optical path of the microscope. Exposure ofthe laser beam is controlled by means of a camera shutter. Lasershots are made every minute. Antithrombotic potency is evaluated

58 Mousa

in real time by microscopic evaluation of vascular occlusion. Thenumber of laser injuries required to induce a thrombus with alength of at least 1.5 times the inner diameter of the vessel istaken as an endpoint.

The antithrombotic activity of several thrombin inhibitors hasbeen compared to unfractionated heparin using the laser-inducedthrombosis model. Each inhibitor was administered intravenouslyvia one of the tail veins and allowed to circulate for 5 min priorto the initiation of the laser-induced lesions. Saline-treated con-trol rats required an average of three laser shots to reach anendpoint. Each thrombin inhibitor produced a dose-dependentantithrombotic effect in this model. In comparing the dose ofeach agent required to extend the endpoint to six laser shots,heparin was observed to be the most potent antithromboticagent (0.08 μmol/kg), followed by Ac-(D)-Phe-Pro-boroArg-OH (0.154 μmol/kg) and then hirudin (0.28 μmol/kg). Consis-tent with the results obtained with these agents in the rabbit jugu-lar vein stasis thrombosis model, D-Me-Phe-Pro-Arg-H exhib-ited the weakest effects in the laser-induced thrombosis model(2 μmol/kg).


In this model, thrombus formation in rat or rabbit mesentericarterioles or venules is induced by laser-mediated thermal injuryto the vascular wall. The procedure can be performed in normalor pretreated (i.e., induced arteriosclerosis or adjuvant arthritis)animals. In this model, thrombus formation is mediated by adual mechanism of platelet adhesion to the injured endothelialvessel wall and ADP-induced platelet aggregation. Most likely,ADP is released by erythrocytes that are lysed by the laser, basedon the observation that erythrocyte hemoglobin strongly absorbsthe frequencies of light emitted by the laser beam. A secondaryaggregation stimulus following the release of ADP is mediated bythe platelets themselves.

2.6.2. Procedure

2.6.2.1. Equipment 1. 4 W argon laser (Spectra Physics, Darmstadt, FRG) witha wave length of 514.5 nm; energy below the objective of15 mW; duration of exposure, 1/30 or 1/15 s.

2. Microscope ICM 405, LD-Epipland 40/0.60 (Zeiss,Oberkochen, FRG)

3. Video camera (Sony, Trinicon tube)4. Recorder (Sony, U-matic 3/4′′)5. Videoanalyzer to determine blood flow velocity



Male Sprague-Dawley, spontaneously hypertensive stroke-proneWistar or Lewis rats with adjuvant-induced arthritis weighing150–300 g are used. Alternatively, New Zealand rabbits with arte-riosclerosis induced by cholesterol feeding for 3 months are used.Animals receive test compound by oral, i.v., i.p. or s.c. admin-istration. Control animals are treated with vehicle alone. Priorto thrombus induction, the animals are pretreated by s.c. injec-tion of 0.1 mg/kg atropine sulfate solution and anaesthetizedby i.p. administration of 100 mg/kg ketamine hydrochloride and4 mg/kg xylazine.

Thrombus formation is induced 15, 30, 60, or 90 min post-dosing. The procedure is carried out in arterioles or venules 13 ±1 μm in diameter of the fat-free ileocaecal portion of the mesen-tery. During the procedure, the mesenterium is superfused witha physiological saline solution or degassed paraffin liquid (37◦C).The ray of the argon laser is guided into the inverted ray pathof the microscope by means of a ray adaptation and adjustingdevice. The frequency of injury is 1 per 2 min. The exposure timefor a single laser shot is 1/30 or 1/15 s. The number of injuriesnecessary to induce a defined thrombus is recorded. All thrombiformed during the observation period with a minimum length of13 μm or an area of at least 25 μm2 are evaluated. The procedureis photographed using a video system.

2.6.2.3. StandardCompounds

• acetylsalicylic acid (10 mg/kg, per os)• pentoxifylline (10 mg/kg, per os)

For a detailed description and evaluation of various agentsand mechanisms, please refer to the following references: Arforset al. (113); Herrmann (114); Seiffge and Kremer (115, 116);Seiffge and Weithmann (117); and Weichert (118).

2.6.3. Evaluation The number of laser shots required to produce a defined throm-bus is determined. Mean values and SEM are calculated. Resultsare typically presented in graph form.

2.7. PhotochemicalInduced Thrombosis


In 1977, Rosenblum and Sabban (119) reported that ultravio-let light can produce platelet aggregation in cerebral microvesselsof the mouse after intravascular administration of sodium fluores-cein, and demonstrated that in contrast to heparin, aspirin andindomethacin prolonged the time to first platelet aggregation. Adetailed study by Herrmann (114) demonstrated that scavengersof singlet oxygen, but not hydroxyl radicals, inhibited platelet

60 Mousa

aggregation induced by photochemical reaction. The investiga-tors postulated that excitation of intravascular fluorescein resultsin the production of singlet oxygen, which damages endothelialcells and leads to platelet adhesion and aggregation.

2.7.2. Procedure Studies are performed in mesenteric arteries 15–30 μm in diame-ter in anesthetized rats. After i.v. injection of 0.3 ml of fluoresceinisothiocyanate–dextran 70 (FITC–dextran; 10%) (Sigma), arteri-oles are exposed to ultraviolet light (excitation, 490 nm; emission,510 nm).

2.7.3. Evaluation Thrombus formation is quantitated by determining the timebetween onset of excitation and appearance of the first plateletaggregate adhering to the vessel wall.

2.7.4. CriticalAssessmentof the Method

In contrast to other thrombosis induction methods, photo-chemically induced thrombosis is amenable to use in small ani-mals. Thrombi are composed primarily of platelets, however, theprimary target of the photochemical insult is endothelial cellsthrough induced oxygen radical damage.


Matsuno et al. (120) reported a method to induce thrombosisin the rat femoral artery by means of a photochemical reactionafter injection of a fluorescent dye (Rose Bengal, 10 mg/kg i.v.)followed by transillumination with a filtered xenon lamp (wavelength, 540 nm). Blood flow was monitored by a pulsed Dopplerflow meter. Occlusion was achieved after approximately 5–6 min.Pretreatment with heparin prolonged the time required to inter-rupt the blood flow in a dose-dependent manner. This model hasalso been used to study the thrombolytic mechanisms of t-PA.For a comparative analysis of hirudin in various models, see Justet al. (45).

2.8. ForeignSurface-InducedThrombosis

The presence of foreign materials in the circulation results in acti-vation of the coagulation and platelet systems. A variety of pro-thrombotic surfaces have been used for the development of exper-imental animal thrombosis models. In contrast to many otherthrombosis models, thrombosis induced by foreign surfaces doesnot presuppose endothelial damage.

2.8.1. Wire Coil-InducedThrombosis

2.8.1.1. Purposeand Rationale

This classical method of producing thrombosis is based on theinsertion of wire coils into the lumen of blood vessels. The modelwas first described by Stone and Lord (121) using the dog aortaand was further modified for use in arterial coronary vessels ofopened-chest dogs. The formation of thrombotic material around


the coil is reproducible and can be easily standardized for phar-macological studies (48, 122, 123).

The use of this model in venous vessels was described byKumada et al. (124). Venous thrombosis is produced in rats byinsertion of a stainless steel wire coil into the inferior caval vein.Platelets and plasmatic coagulation are activated on the wire coil.Thrombus formation on the wire is quantitated by measuring theprotein content of the isolated thrombotic material. The kinet-ics of thrombus formation show an increase in weight and pro-tein content within the first 30 min of insertion, followed bya period of steady state flux between thrombus formation andendogenous thrombolysis, leading to a level protein content ofthrombi starting at 1 h and lasting up to 48 h after implantation.The incidence of thrombosis in untreated control animals in thismodel is 100%. The model is used to evaluate antithrombotic andthrombolytic properties of test compounds in an in vivo model ofvenous thrombosis in rats.

2.8.1.2. Procedure Male Sprague-Dawley rats weighing 260–300 g receive test com-pound, or vehicle as a control, by oral, i.v. or i.p. administration.At the end of absorption (i.v., 1 min; i.p., 30 min; p.o., 60 min),the animals are anesthetized by i.p. injection of 1.3 g/kg of ure-thane. Through a midline incision the caudal caval vein is exposedand a stainless steel wire coil (Zipperer R©, size 40; Zdarsky ErlerKG, München) is inserted into the lumen of the vein just belowthe left renal vein branching by gently twisting the wire towardthe iliac vein. The handle of the carrier is cut off so as to holdthe back end of the wire at the vein wall. The incision is suturedand the animal is placed on its back on a heating pad (37◦C). Thewound is reopened after 2 h and the wire coil with the thrombuson it is carefully removed and rinsed with a 0.9% saline solution.The thrombotic material is dissolved in 2 ml of alkaline sodiumcarbonate solution (2% Na2CO3 in 0.1 N NaOH) in a boilingwater bath for 3 min. The protein content is determined in 100 μlaliquots by the colorimetric method of Lowry (Fig. 2.4).

Thrombolysis. In addition to the procedure described above,a thrombolytic test solution is continuously infused through apolyethylene catheter inserted into the jugular vein. Ninety min-utes after wire implantation, the test compound or the vehi-cle (control) is infused for up to 2.5 h. The wire coil is thenremoved and the protein content of the thrombus is deter-mined. Using this model, Bernat et al. (125) demonstrated thefibrinolytic activity of urokinase and SK–human plasminogencomplex.

2.8.1.3. Evaluation 1. Thrombosis incidence = number of animals in each dosagegroup that develop thrombi as compared to the vehiclecontrol.

62 Mousa

Fig. 2.4. Schematic diagram of the canine femoral artery copper coil model of thrombolysis. A thrombogenic copper coilis advanced to either femoral artery via the left carotid artery. By virtue of the favorable anatomical angles of attachment,a hollow polyurethane catheter advanced down the left carotid artery nearly always enters the descending aorta, andwith further advancement, either femoral artery without fluoroscopic guidance. A flexible, teflon-coated guidewire isthen inserted through the hollow catheter and the latter is removed. A copper coil is then slipped over the guidewireand advanced to the femoral artery (see inset). Femoral artery flow velocity is measured directly and continuously witha Doppler flow probe placed just proximal to the thrombogenic coil and distal to a prominent sidebranch, which is leftpatent to dissipate any dead space between the coil and the next proximal sidebranch. Femoral artery blood flow declinesprogressively to total occlusion over the next 10–12 mm after coil insertion.

2. The mean protein content [mg] of thrombotic material ineach dosage group as compared to the vehicle control isdetermined. Percent change in protein content is calculatedrelative to control.

3. Statistical significance is assessed by means of the unpairedStudent’s t-test.

2.8.2. EversionGraft-InducedThrombosis


The eversion graft model of thrombosis in the rabbit artery wasfirst described by Hergrueter et al. (126) and later modified by


Jang et al. (127, 128) and Gold et al. (15). A 4- to 6-mm seg-ment of the rabbit femoral or the dog left circumflex artery isexcised, everted, and then re-implanted into the vessel by end-to-end anastomosis. After restoration of blood flow, a platelet-richocclusive thrombus forms rapidly leading to complete occlusionof the vessel. The rabbit model described here uses a carotid graftinserted into the femoral graft to avoid vasoconstriction, whichoften occurs in the inverted femoral segments.

2.8.2.2. Procedure In anaesthetized New Zealand white rabbits, the right carotidartery is exposed. After double ligation, a 3-mm segment of theartery is excised, everted, and immersed in pre-warmed (37◦C)isotonic saline. The right femoral artery is exposed and occludedby means of a double occluder (2-cm distance). The femoralartery is transected and the everted graft from the carotid arteryis inserted by end-to-end anastomosis using 12 sutures and 9-0nylon (Prolene; Ethicon, Norderstedt, Germany) under a surgicalmicroscope (Wild M650; Leitz, Heerbrugg, Switzerland). Perfu-sion of the graft is measured by means of an ultrasonic flow meter(Model T106; Transonic, Ithaca, NY, USA). The flow probe ispositioned 2 cm distal from the graft. After a stabilization periodof 15 min, the test substance is administered i.v. through thecatheterized right jugular vein. Ten minutes after administrationof the test compound, the vessel clamps are released and bloodflow is monitored by the flow meter for 120 min.

Arterial blood is collected from the left carotid artery at base-line (immediately before administration of test compound), and10, 60, and 120 min after administration.

2.8.2.3. Evaluation 1. Time until occlusion = time between restoration of vesselblood flow and occlusion of the vessel, as indicated by a flowof less than 3.0 ml/min.

2. Patency = time during which perfusion of the graft ismeasured relative to an observation period of 120 min afteradministration of test compound.

3. Time until occlusion and patency are expressed as medianand inter-quartile range/2 (IQR/2). Significant differences(P<0.05) are calculated by the nonparametric Kruskal–Wallistest.

2.8.2.4. CriticalAssessmentof the Method

The eversion graft is very thrombogenic, although technicallydifficult and time consuming. The resultant deep occlusivethrombi can be prevented only by intra-arterial administrationof thrombolytics or aggressive antithrombotic treatments, suchas high doses of recombinant hirudin or PEG-hirudin. Becausethe initiating surface is a non-endothelial tissue containing tissuefactor and collagen, both the coagulation and the blood plateletsystems are activated.

64 Mousa

2.8.2.5. Modificationsof the Method

Gold et al. (15) described a modification of this model in partiallyobstructed left circumflexed coronary arteries of thoracotomizeddogs. The combination of reduced blood flow due to the con-strictor and an abnormal non-endothelial surface produces totalthrombotic occlusion within 5 min.

Models that use a catheter to induce vessel wall damage ofboth arteries and veins have been reported (63, 129–131). Suchmodels in the arterial system mimic potential injuries inducedby angioplasty. In these models, the endothelium is damagedeither by rubbing the catheter across the luminal surface of thevessel or by air desiccation. Inflation of the balloon and theinduction of partial stasis in the area of damage produce addi-tional injury. By this procedure, vessel wall collagen, elastic tis-sues, and tissue thromboplastin are exposed to the circulatingblood. Such models are typically carried out in rabbits or largeranimals due to size considerations for both the vessel and thecatheter.

In these models, the formation of thrombi has been detectedin a number of ways. Measurement of flow by a distally placedflow meter has been reported (130). A decrease in vessel tem-perature measured distally to the site of injury is reflective of adecrease in blood flow through the segment and the formationof a thrombus. Deposition of radiolabeled platelets at the site ofinjury and measurement of thrombus wet weight have also beenused.

Platelets appear to play an important role in the formationof thrombi at sites where the endothelium is damaged (132).Platelets may also play a key role in the initiation of the restenoticprocess following angioplasty (133). These models, therefore,provide the opportunity to assess the pharmacologic effects ofagents capable of modulating either acute platelet function or thecoagulation system that may be useful as adjunctive treatments inangioplasty. It has been demonstrated that both the platelet andthe clotting systems are activated by arterial intervention (133,134) and with this model, it has been shown that heparin andhirudin are both capable of inhibiting initial thrombosis. In addi-tion, these models have also been used to assess the inhibition ofre-thrombosis following lysis of the initial clot (130).

2.8.3. Arteriovenous (AV)Shunt Thrombosis


A method for the direct observation of extracorporeal throm-bus formation was introduced by Rowntree and Shionoya(135). Very early studies using this model provided evidencethat anticoagulants like heparin and hirudin inhibit thrombusdevelopment in AV shunts. Today, AV shunt thrombosis modelsare often used to evaluate the antithrombotic potential of newcompounds in different species including rabbits (136), rats


(137), pigs (138), dogs and cats (139), and non-human pri-mates (140).

2.8.3.2. Procedure Rats are anaesthetized and fixed in a supine position on atemperature-controlled heating plate to maintain body temper-ature. The left carotid artery and the right jugular vein arecatheterized with short polyethylene catheters. The catheters arefilled with isotonic saline solution and clamped. The two endsof the catheters are connected with a 2-cm glass capillary withan internal diameter of 1 mm. This glass capillary provides thethrombogenic surface. At a defined time after administration oftest compound, the clamps that are occluding the AV shunt areopened.

Patency of the shunt is measured indirectly using a NiCrNi-thermocouple fixed distal to the glass capillary. When blood isflowing, the temperature rises from room temperature to bodytemperature. By comparison, decreased temperature indicates theformation of an occluding thrombus. Temperature is measuredcontinuously over 30 min after the opening of the shunt.


It has been shown by Best et al. (139) that thrombi formed inthe AV shunt are to a great extent white arterial thrombi. Thismight be due to the high pressure and shear rate inside the shunts,causing thrombi to be more arterial in character (14).

2.8.4. Thread-InducedVenous Thrombosis


Compared to the arterial system, the development of thrombosismodels in venous blood vessels tends to be more difficult in termsof reproducibility and variability (14). Complete stasis togetherwith a thrombogenic stimulus (Wessler-type) has been used bya number of investigators to evaluate the effects of test com-pounds on venous thrombosis. Hollenbach et al. (141) developeda rabbit model of venous thrombosis by introducing cottonthreads into the abdominal vena cava of rabbits. The cottonthread serves as a thrombogenic surface, and the thrombus thatforms around it reaches a maximum mass after 2–3 h. The pro-longed non-occlusive character of thrombogenesis in this modelenables studies that focus on the progression of thrombus forma-tion rather than initiation. Thus, conditions more closely resem-ble the pathophysiology of thrombosis in humans, because bloodcontinues to flow throughout the experiment (14).

2.8.4.2. Procedure Rabbits weighing 2.5–3.5 kg are anaesthetized by isofluoraneinhalation anesthesia, and a polyethylene catheter is insertedinto the left carotid artery. A polyethylene tube (PE 240; innerdiameter, 1.67 mm) 14 cm in length is filled with isotonic saline,and a copper wire affixed to five fixed cotton threads (6 cm inlength) is inserted into the tube (after determination of the netweight of the cotton threads). A laparotomy is performed and the

66 Mousa

vena cava and vena iliac are dissected free from surrounding tis-sue. The test compound is administered by an intragastric tubefor 60 min (depending on the results of ex vivo analysis) prior toinitiation of thrombus formation. Blood samples are analyzed 60,90, 120, 150, and 210 min after oral administration of the testcompound.

Thrombus formation is induced by inserting the thrombosiscatheter into the caval vein via the vena iliaca (7 cm). The cop-per wire is pushed forward 3 cm to release the cotton threadsinto the vessel lumen. After thrombus initiation (150 min afterinitiation), the caval segment containing the cotton threads andthe developed thrombus is removed, opened longitudinally, andthe content is blotted onto filter paper. After weighing the cottonthreads with thrombus, the net dry thread weight is subtracted todetermine the corrected thrombus weight.

2.8.4.3. Evaluation 1. Corrected thrombus weight after blotting the thrombus onfilter paper and subtraction of the net dry weight of the cot-ton thread.

2. Mean arterial blood pressure (MAP).3. aPTT, HepTest, anti-FIIa, and anti-FXa activities.


The cotton thread-induced thrombus is composed of fibrintogether with tightly aggregated and distorted erythrocytes, sim-ilar to human deep vein thrombosis structure. Non-occlusivethrombus formation in this model has been successfully inhibitedby heparins, prothrombinase complex inhibitors, and thrombininhibitors (141, 142).

2.8.4.5. Modificationsof the Method

In addition to the originally described method, it is possible tomeasure blood flow by means of an ultrasonic flow probe attacheddistal to the position of the cotton threads on the vein.

2.8.5. ThrombusFormation onSuperfused Tendon


In all models that include vessel wall damage, blood comes incontact with adhesive proteins of the sub-endothelial matrix, i.e.,von Willebrand factor, collagen, fibronectin, laminin, and others.Gryglewski et al. (143) described an in vivo method in which theblood of a non-anesthetized animal is exposed ex vivo to a for-eign surface consisting mainly of collagen. The foreign surface is apart of the tendon of another animal species. After superfusion ofthe tendon, blood is re-circulated to the non-anesthetized animal.This method enables the quantitation of antiplatelet potencybased on the formation of platelet thrombi on the surface of thetendon or aortic strips from atherosclerotic rabbits.


2.8.5.2. Procedure Blood is withdrawn from the carotid artery of anesthetized andheparinized cats using a roller pump at a speed of 6 ml/min.After a passage through a warmed jacket (37◦C), blood is sep-arated into two streams, each flowing at a speed of 3 ml/min,that superfuse in parallel two twin strips of the central part ofthe longitudinally cut rabbit Achilles tendon (30×3 mm). Aftersuperfusing the tendon strips, the blood is allowed to drip intocollectors and return to the venous system of the animal by grav-ity through the left jugular vein. The tendon strips are freely sus-pended in the air. The upper ends are tied to an auxotonic leverof a smooth muscle/heart Harvard transducer; the lower ends areloaded with a weight (1–2 g) to keep the lever with its counter-weight in a neutral position. When superfused with blood, thestrips become covered with clots, changing the weight of thestrips. Weight changes are continuously recorded. After a con-trol period of 30 min, the formed thrombi are gently removedand fixed in formalin for histological examination. The strips aresuperfused with Tyrode solution, and the animals are injectedwith antithrombotic (test) compound. After 10 min, blood super-fusion is repeated for another 30 min.

2.8.5.3. Evaluation The ratio of weight increase of the strips after drug treatmentrelative to before drug treatment is taken as an index of anti-aggregatory activity.

2.9. Stasis-InducedThrombosis (WesslerModel)


The Wessler model is a classic method of inducing venous throm-bosis in animals. Wessler (3, 144–148) combined local venousstasis with hypercoagulability produced by injection of humanor dog serum into the systemic circulation of dogs or rabbits.The jugular vein of these animals is occluded by clamps 1 minafter injection of the procoagulatory stimulus into the circulation.Within a few minutes after clamping, a red clot is formed in theisolated venous segment. Fareed et al. (149) summarizes the vari-ety of substances that can be used as procoagulatory stimuli in thismodel. Aronson and Thomas (150) found an inverse correlationbetween the duration of stasis and the amount of hypercoagula-tion agent used to produce the clot.

2.9.2. Procedure Anaesthetized rabbits are fixed in a supine position on atemperature-controlled (37◦C) heating table. Following cannu-lation of both carotid arteries (the left in a cranial direction)and the right vena femoralis, segments (2 cm in length) of thetwo external jugular veins are exposed and isolated between

68 Mousa

two loose sutures. Calcium thromboplastin (0.3 ml/kg) (Sigma;Deisenhofen, Germany, FRG) is administered via the left carotidartery. Meticulous care is taken to maintain a standard injectiontime of 30 s followed by injection of 0.5 ml of physiological salinewithin 15 s. Both jugular vein segments are occluded 45 s later bydistal and proximal sutures. Stasis is maintained for 30 min. Bloodsamples are taken immediately before occlusion and 30 s beforethe end of stasis. After excision, the occluded vessel segments areplaced on a soaked sponge and opened by longitudinal incision.

2.9.3. Evaluation The size of the clots is assessed using a scoring system: 0, bloodonly; 1, very small clot piece, filling up to 1/4 of the vessel; 2,larger clot, filling up to 1/2 of the vessel; 3, very large clot, fill-ing up to 3/4 of the vessel; 4, a single, large clot that fills thewhole vessel. The scores for the left and the right jugular veins areadded to arrive at a thrombus size value for each animal. Throm-bus weight is also measured after blotting the thrombus on filterpaper.

Thrombus score is expressed as a median (minimum–maximum). Thrombus weight is expressed as mean ±SEM. Forthe statistical evaluation of antithrombotic effects, the nonpara-metric U-test of Mann and Whitney (thrombus score) or Stu-dent’s t-test for unpaired samples (thrombus weight) is used.Significance is expressed as P < 0.05.


Because of its static design, Breddin (151) described the Wesslermodel as the retransformation of an in vitro experiment into avery artificial test situation. One of the major drawbacks of theWessler model is that it is relatively independent of platelet func-tion and hemodynamic changes that largely influence thrombusformation in vivo. However, the model has been shown to be veryuseful for evaluating the antithrombotic effects of compounds likeheparin and hirudin.


There are a number of different procoagulant agents, such ashuman serum, Russel viper venom, thromboplastin, thrombin,activated prothrombin complex concentrates, and FXa, that havebeen used to induce thrombosis in this model (149, 150). Thesensitivity and accuracy of the model can be improved by injectingiodinated fibrinogen into the animals before injecting the throm-bogenic agent and then measuring specific radioactivity of theclot.

A general drawback of the Wessler model is the static natureof venous thrombus development. To overcome this problem,some investigators have developed more dynamic models thatincorporate reperfusion of the occluded vessel segments afterclot development. Depending on the time of administration oftest compound (pre- or post-thrombus initiation), the effect on


thrombus growth and fibrinolysis can be evaluated. Levi et al.(152) have used this model to assess the effects of a murine mon-oclonal anti-human PAI-1 antibody, and Biemond et al. (153)compared the effects of thrombin and FXa inhibitors to a low-molecular weight heparin using a modified Wessler model.

Venous reperfusion model. New Zealand white rabbits weigh-ing 2.5 kg are anesthetized with 0.1 ml of atropine, 1.0 mg/kgof diazepam, and 0.3 ml of Hypnorm (Duphar; 10 mg/mlfluanisone and 0.2 ml fentanyl). Anesthesia is maintained with4 mg/kg i.v. thiopental. The carotid artery is cannulated afterexposure through an incision in the neck. The jugular vein is dis-sected free from tissue and small side branches are ligated over adistance of 2 cm. The vein is clamped proximally and distally toisolate the vein segment. Citrated rabbit blood (from another rab-bit) is mixed with 131I-fibrinogen (approximately 25 mCi/ml),and then 150 μl of radiolabeled blood is aspirated in a 1-mlsyringe containing 25 μl of thrombin (3.75 IU) and 45 μl of0.25 M CaCl2. An aliquot (200 μl) of the clotting blood isimmediately injected into the isolated segment. Thirty minutesafter clot injection, the vessel clamps are removed and bloodflow is restored. 125I-fibrinogen (approximately 5 μCi) is injectedthrough the cannula in the carotid artery (in the case of fibrinoly-sis studies, this is immediately followed by injection of 0.5 mg/kgrecombinant t-PA). For each dosage group, four thrombi are ana-lyzed. The extent of thrombolysis is assessed by measuring 131I-fibrinogen remaining in the clot (relative to initial clot radioac-tivity). Comparison of 125I levels in blood and thrombus is ameasure of the extent of thrombus growth. Thrombus lysis andextension are monitored 60 or 120 min after thrombus forma-tion and are expressed as a percentage of the initial thrombusvolume. Statistical analysis is carried out using variance analysisand the Newman–Keuls test. Statistical significance is expressedas P<0.05.

2.10. DisseminatedIntravascularCoagulation (DIC)Model


Widely used in rats and mice, the DIC model is a modelof systemic thrombosis induced by tissue factor, endotoxin(lipopolysaccharide), or FXa (154–156). After systemic admin-istration of a thrombogenic stimulus, studies can be performedwith or without mechanical vena caval stasis. When stasis is used,the major parameter is thrombus mass; when stasis is not used,the parameters are primarily fibrin degradation products, fibrino-gen, platelet count, prothrombin time (PT), and activated par-tial thromboplastin time (aPTT). Given the many and varied

70 Mousa

parameters that are measured when stenosis is not used, post-experimental analysis can be time consuming and technicallydemanding. Although rodents are useful as a primary efficacymodel, limitations in drawing multiple blood samples over thecourse of the experiment and differences in activity of at leastsome FXa inhibitors in human as compared to rat plasma invitro require that compounds be further characterized in moreadvanced in vivo models of thrombosis.

2.11. MicrovascularThrombosis inTrauma Models


Successful re-plantation of amputated extremities is dependentto a large degree on maintaining the microcirculation. A num-ber of models have been developed in which blood vessels aresubjected to crush injury with or without vascular avulsion andsubsequent anastomosis (157–159). In the model of Stockmans(159), both femoral veins are dissected from the surrounding tis-sue. A trauma clamp, which has been adjusted to produce a pres-sure of 1,500 g/mm2, is positioned parallel to the long axis ofthe vein. The anterior wall of the vessel is grasped between thewalls of the trauma clamp and the two endothelial surfaces arerubbed together for a period of 30 s as the clamp is rotated. For-mation and dissolution of platelet-rich mural thrombi are moni-tored over a period of 35 min by transillumination of the vessel.By using both femoral veins, the effect of drug therapy can becompared to control in the same animal, minimizing intra-animalvariations.

The models of Korompilias (158) and Fu (157) examine theformation of arterial thrombosis in rats and rabbits, respectively.In these models, either the rat femoral artery or the rabbit centralear artery is subjected to a standardized crush injury. The ves-sels are subsequently divided at the midpoint of the crushed areaand then anastomosed. Vessel patency is evaluated by milking thevessel at various time points post-anastomosis. These models havebeen used to demonstrate the effectiveness of topical administra-tion of LMWH in preventing thrombotic occlusion of the vessels.Such models, while effectively mimicking the clinical situation,are limited by the necessity of a high degree of surgical skill toeffectively anastomose the crushed arteries.

2.12.CardiopulmonaryBypass Models


Cardiopulmonary bypass (CPB) models have been described inbaboons (160), swine (161), and dogs (162). In each model, thevariables that can affect the hemostatic system such as anesthesia,


shear stress caused by the CPB pumps, and the exposure of plasmacomponents and blood cells to foreign surfaces (i.e. catheters,oxygenators, etc) are comparable to that observed with humanpatients. With these models, it is possible to examine the poten-tial usefulness of novel anticoagulants in preventing thrombo-sis under relatively harsh conditions where both coagulation andplatelet function are altered. The effectiveness of direct throm-bin inhibitors (160), LMWHs (163), and heparinoids (162) hasbeen compared to standard heparin using this model. Endpointshave included the measurement of plasma anticoagulant levels,histological determination of microthrombi deposition in variousorgans, formation of blood clots in components of the extra-corporeal circuit, and the deposition of radiolabeled platelets invarious organs and components of the extracorporeal circuit.These models, therefore, can be used to assess the antithromboticpotential of new agents for use in CPB surgery and also to assessthe biocompatibility of components used to maintain extracorpo-real circulation. The reader is referred to several detailed protocolsand evaluations of this model (3–9, 157–159).

2.13. ExtracorporealThrombosis Models


These models employ passing blood over a section of damagedvessel (or other selected substrate) and recording thrombus accu-mulation on the damaged vessel histologically or by scintigraphicdetection of radiolabeled platelets or fibrin (164). This model isinteresting because the results can be directly compared to in vivodeep arterial injury model (165) results and to results from asimilar extracorporeal model used in humans (166, 167). Dan-gas et al. (166) used this model to characterize the antithrom-botic efficacy of abciximab, a monoclonal antibody-based plateletGPIIb/IIIa inhibitor, after administration to patients undergo-ing percutaneous coronary intervention. They demonstrated thatabciximab reduces both the platelet and the fibrin components ofthe thrombus, thereby providing further insight into the uniquelong-term effectiveness of short-term administration of this drug.Ørvim et al. (167) also used this model in humans to evalu-ate the antithrombotic efficacy of recombinant-tick anticoagulantpeptide (rTAP), but instead of evaluating the compound afteradministration of rTAP to the patient, the drug was mixed withthe blood immediately as it flowed into the extracorporeal cir-cuit prior to flowing over the thrombogenic surface. By chang-ing the thrombogenic surface, they were able to determine thatrTAP was more effective at inhibiting thrombus formation on atissue factor-coated surface compared to a collagen-coated sur-face. These results suggest that optimal antithrombotic efficacyrequires an antiplatelet approach along with an anticoagulant.

72 Mousa

Although this model does not completely represent patholog-ical intravascular thrombus formation, the use of this “humanmodel” of thrombosis may be very useful in developing newdrugs because it directly evaluates the ex vivo antithromboticeffect of a drug in flowing human blood.

2.14. ExperimentalThrombocytopeniaor Leucocytopenia


Intravenous administration of collagen, arachidonic acid, ADP,PAF, or thrombin activates thrombocytes leading to maximalthrombocytopenia within a few minutes. This effect is reinforcedby additional injections of epinephrine. Activation of plateletsleads to intravascular aggregation and temporary sequestrationof aggregates in the lungs and other organs. Depending on thedose of agonist, this experimentally-induced reduction in circu-lating platelets is reversible within 60 min after induction. Fol-lowing administration of PAF (or other agonist), leucocytopeniais induced by the addition of epinephrine. This assay is used to testthe inhibitory capacity of compounds against thrombocytopeniaor leucocytopenia arising from in vivo platelet or leukocyte stim-ulation.

2.14.2. Procedure Male guinea pigs (Pirbright White) (300–600 g), male NMRImice (25–36 g), or Chinchilla rabbits of either sex (2–3 kg) areused. Animals receive test compound or vehicle as a control byoral, i.p., or i.v. administration (Table 2.1). After absorption time(p.o., 60 min; i.p., 30 min; i.v., variable), the marginal vein ofthe ear of rabbits is cannulated and a thrombocytopenia-inducingsubstance (i.e., collagen or arachidonic acid) is injected slowly.Blood is collected from the ear artery.

Guinea pigs, hamsters, or mice are anesthetized with pen-tobarbital sodium (i.p.) and Rompun R© (i.m.) and placed ona temperature-controlled table at 37◦C. The carotid artery iscannulated for blood withdrawal and the jugular vein is can-nulated to administer thrombocytopenia-inducing substance(s),such as collagen+adrenaline, PAF, or thrombin (see Table 2.1).For mice, collagen+adrenaline is injected into a tail vein. Approx-imately 50–100 μl of blood is collected into potassium-EDTA-coated tubes 1 min before (–1), and 1 and 2 min after injectionof the inducer (for guinea pigs and mice), or 5, 10, and 15 min(for rabbits) after inducer. The number of platelets and leuko-cytes is determined within 1 h of withdrawal in aliquots of 10 μlof whole blood using a microcell counter suitable for blood ofvarious animal species.


Table 2.1Experimental thrombocytopenia or leucocytopenia: materials and solutions

Substance used to induce thrombocytopenia/leucocytopenia(i.v. administration) Dose

Rabbits:

Arachidonic acid (Sigma) 1 mg/kgCollagen (Hormonchemie) 30 μg/ml

Mice:Collagen + adrenaline (Hormonchemie) 90 μg/kg + 20 μg/kg

Hamsters:Collagen + adrenaline 50 μg/kg + 10 μg/kg

Guinea pigs:PAF (Paf-acether, Bachem) 0.03–0.04 μg/kg

Thrombin (Hoffman-LaRoche) 60 U/kgAnesthetics:

Pentobarbital sodium (i.p.) 30 mg/kgXylazine (i.m.) 8 mg/kg

Urethane (i.p.) 1.5 g/kgPlatelet analyzer: Sysmex microcellcounter F-800

2.14.3. Evaluation 1. The percentage of thrombocytes (or leukocytes) in vehi-cle control and dosage groups at different time points afterinjection of the inducer is determined relative to the initialvalue (before injection). The values for the controls are setas 100%.

2. Percent inhibition of thrombocytopenia (or leucocytopenia)is calculated in each dosage group relative to the control.



The method of collagen + epinephrine-induced thrombocy-topenia has been widely used to study the phenotypes ofknock-out mice carrying deletions of specific genes implicatedin hemostasis/thrombosis. Recent examples are the Gas 6−/−knockout mouse (168) and mice lacking the gene for the Gz smallG protein (169). The advantages of this method are that it is asimple experimental procedure, and a small volume of blood isrequired. In general, application of the method in small animals(mice, hamsters) requires small amounts of test compound. Thismodel is a useful first step in assessing the in vivo antithromboticefficacy of antiplatelet drugs.

74 Mousa

2.15.Collagenase-InducedThrombocytopenia


Intravenous administration of the proteolytic enzyme collage-nase leads to the formation of endothelial gaps and exposure ofdeeper layers of the vessel wall. This type of vascular endothe-lial injury mainly triggers thrombus formation through the acti-vation of platelets by contact with the basal lamina. As a conse-quence, thrombocytopenia is induced, which is maximal within5–10 min following collagenase injection and reversible within30 min after induction. This model is used as an alternativeto the model described above to test the inhibitory capacity ofcompounds against thrombocytopenia in a model of collagenase-induced thrombocytopenia in rats.

2.15.2. Procedure Male Sprague-Dawley rats weighing 260–300 g are used. The ani-mals receive test compound, or vehicle as a control, by oral, i.p.,or i.v. administration. After absorption time (i.p., 30 min; p.o.,60 min; i.v., variable), rats are anesthetized with pentobarbitalsodium (i.p.) (see Table 2.2). One carotid artery is cannulatedfor blood withdrawal and one jugular vein is cannulated for injec-tion of inducer. The animals receive an i.v. injection of heparin,and then 20 min later, approximately 100 μl of blood is collected(initial value). Ten minutes later, a thrombocytopenia-inducingsubstance (collagenase) is administered intravenously.

Five, ten, twenty, and thirty minutes after the injection of col-lagenase, blood samples (approximately 100 μl each) are collectedinto potassium–EDTA-coated tubes. The number of platelets isdetermined in 10 μl aliquots of whole blood within 1 h afterblood withdrawal using a microcell counter. For additional detailsof this method, see Volkl and Dierichs (170).

2.15.3. Evaluation 1. The percentage of platelets in vehicle control and dosagegroups at the different times following injection of collagen-

Table 2.2Collagenase-induced thrombocytopenia: materials andsolutions

Anesthetic: pentobarbital sodium (i.p.) 60 mg/kg

Heparin (Liquemin R©) (i.v.) 500 U/kgThrombocytopenia induction: collagenase (i.v.)

(E.C. 3.4.24.3; Boehringer Mannheim)10 mg/ml/kg

Platelet analyzer: Sysmex microcellcounter F-800


ase is determined relative to the initial value for each group.The values for the controls are set as 100%.

2. Percent inhibition of thrombocytopenia is calculated in eachdosage group relative to the control.


2.16. ReversibleIntravitalAggregationof Platelets


Isotopic labeling of platelets can be used to monitor plateletaggregation and desegregation in vivo. ADP, PAF, arachidonicacid, thrombin, and collagen are known to induce platelet aggre-gation. In this model, labeled platelets are continuously moni-tored in the thoracic (A) and abdominal (B) region of test ani-mals. Administration of aggregation-promoting agents producesan increase in radioactivity in A and a decrease in radioactivity inB. This method assumes that platelets aggregate within the vas-cular system and accumulate in the pulmonary microvasculature.This in vivo method can be used to evaluate the platelet anti-aggregatory properties of test compounds.

2.16.2. Procedure

2.16.2.1. Preparationof Labeled Platelets

Blood is obtained from rats by cardiopuncture. After centrifuga-tion at 240×g for 10 min, platelet-rich plasma (PRP) is trans-ferred to a clean tube and re-suspended in calcium-free Tyrodesolution containing 250 ng/ml prostaglandin E1 (PGE1). Thesuspension is subjected to centrifugation at 640×g for 10 min.The supernatant is discarded and the sediment is suspendedby gentle shaking in calcium-free Tyrode solution containing250 ng/ml PGE1. 51Cr is added to 1 ml of the platelet suspen-sion. Following a 20-min incubation period at 37◦C, the suspen-sion is again subjected to centrifugation at 640×g for 10 min. Thesupernatant is removed and the labeled platelets are re-suspendedin 1 ml of calcium-free Tyrode solution containing 250 ng/mlPGE1.


Male Sprague-Dawley or stroke-prone spontaneously hyperten-sive rats weighing 150–300 g are used. The animals are anaes-thetized with pentobarbital sodium (30 mg/kg, i.p.). Follow-ing tracheotomy, the vena femoralis is exposed and cannulated.The labeled platelets are administered via the cannula. Circulat-ing platelets are monitored continuously in the thoracic (A) andabdominal (B) region. Counts are collected using a dual channelgamma spectrometer (Nuclear Enterprise 4681) integrated witha microcomputer (AM 9080A). One hour after administration of

76 Mousa

labeled platelets (or when counts in A and B have stabilized), anaggregation-promoting agent (i.e., ADP, PAF, arachidonic acid,thrombin, or collagen) is administered twice by i.v. injection. Onehour is allowed to elapse between each i.v. injection.

The test compound is administered 2 h after platelet injectionconcurrently with the fourth administration of the aggregatingagent. Thirty minutes (for ADP, PAF, arachidonic acid, or throm-bin), or one hour (for collagen) after administration of test com-pound, another injection of aggregating agent is administered.This injection serves as an additional control or to determine thelong-term efficacy of a test compound.

2.16.2.3. StandardCompound

PGI2 (prostacyclin).

2.16.3. Evaluation 1. The microcomputer continuously collects information aboutaggregation and desegregation of labeled platelets.

2. The following parameters are recorded:A: counts over the thoraxB: counts over the abdomenDifference: A–BRatio: A/B

3. The time course of response is represented by a curve. Thearea under the curve is calculated using a specific computersoftware program.

4. Statistical significance is calculated using the Student’s t-test.

2.16.4. Modificationof the Method

Oyekan and Botting (171) described a method for monitoringplatelet aggregation in vivo in rats using platelets labeled withindium3+ oxine in which the increase in radioactivity in the lungafter injection of ADP or collagen was recorded.

Smith et al. (172) continuously monitored the intra-thoraciccontent of intravenously injected 111indium-labeled platelets inanesthetized guinea pigs using a microcomputer-based system.

3. Animal Modelsof Bleeding

3.1. Subaqueous TailBleeding Timein Rodents


Blood vessel damage results in the formation of a hemostatic plug,a process that involves several different mechanisms, includingvascular spasm, formation of a platelet plug, blood coagulation,and growth of fibrous tissue into the blood clot. A diagnosticparameter for specifically assessing defects of the hemostatic


system and the influence of drugs on hemostasis is the lengthof time that it takes for bleeding to stop from a standard incision,the so-called bleeding time.

Bleeding-time measurements in animals are used to evalu-ate the hemorrhagic properties of antithrombotic drugs. Tran-section of the tail of a rodent was first established by Döttl andRipke (173), and today it is commonly used in experimentalpharmacology.

3.1.2. Procedure Anaesthetized rats are fixed in a supine position on a temperature-controlled (37◦C) heated table. Following catheterization of acarotid artery (for measurement of blood pressure) and a jugu-lar vein, test compound is administered. After a defined latencyperiod, the tail of the rat is transected with a razor blade mountedon a self-constructed device at a distance of 4 mm from the tip ofthe tail. Immediately after transection, the tail is immersed into abath filled with isotonic saline solution (37◦C).

3.1.3. Evaluation The time until bleeding stops is determined within a maximumobservation time of 600 s.


There are numerous variables that can influence bleeding timemeasurements in the rodent, as discussed in detail by Dejanaet al. (174), including the position of the tail (horizontal or verti-cal), the environment (air or saline), temperature, anesthesia, andmethod of injury (i.e., Simplate method, transection). These vari-ables contribute to differences in results reported for compoundslike aspirin and heparin analyzed under different assay conditions(175, 176). Furthermore, it is impossible to transect exactly oneblood vessel, because the transected tail region consists of a fewmajor arteries and veins, which mutually interact.

3.2. Arterial BleedingTime in Mesentery


Arterial bleeding is induced by micro-puncture of small arteriesin the area supplied by the mesenteric artery. Bleeding is arrestedin viable blood vessels by the formation of a hemostatic plug dueto the aggregation of platelets and fibrin formation. In this test,compounds can be evaluated for their ability to inhibit thrombusformation, and thus prolong arterial bleeding time. This test isused to assess agents that interfere with primary hemostasis insmall arteries.

3.2.2. Procedure Male Sprague-Dawley rats weighing 180–240 g receive test com-pound, or vehicle as a control, by oral, i.p., or i.v. administration.After absorption time (i.p., 30 min; p.o., 60 min; i.v., variable),animals are anesthetized by i.p. injection of 60 mg/kg pentobar-bital sodium. Rats are placed on a temperature-controlled table at37◦C.

78 Mousa

The abdomen is opened by a midline incision and the mesen-tery is lifted to display the mesenteric arteries. The mesenteryis draped over a plastic plate and superfused continuously withTyrode solution maintained at 37◦C. Bleeding times are deter-mined on small mesenteric arteries (125–250 μm external diame-ter) at the junction of the mesentery with the intestines. Adiposetissue surrounding the vessel is carefully cut with a surgicalblade.

Arteries are punctured with a hypodermic needle (25 gauge;16×5/10 mm). The bleeding time of the mesenteric blood ves-sels is observed through a microscope at a magnification of 40×.The time in seconds is determined from the time of puncture untilbleeding is arrested by the formation of a hemostatic plug.

3.2.3. Evaluation 1. Mean bleeding times [s] are determined for each dosagegroup (4–6 animals/group, 4–6 punctures per animal) andcompared to the respective control.

2. Statistical significance assessed by means of the unpaired Stu-dent’s t-test.

3. Prolongation of bleeding time in each dosage group is cal-culated relative to the vehicle control.

For additional details on the method and its use in evaluatingvarious mechanisms or agents, refer to Butler et al. (177), Dejanaet al. (174), and Zawilska et al. (178).

3.3. TemplateBleeding Time


Template bleeding time is used to detect abnormalities of pri-mary hemostasis due to deficiencies in the platelet or coagula-tion system by way of a standardized linear incision in the skin(for human). This method has been modified with the develop-ment of a spring-loaded cassette with two disposable blades (Sim-plate II, Organon Teknika, Durham, NC). These template devicesensure reproducibility in the length and depth of dermal incisions.Forsythe and Willis (179) described a modification that enablesthe Simplate technique to be used to analyze bleeding time in theoral mucosa of dogs.

3.3.2. Procedure The dog is positioned in sternal or lateral recumbence. A strip ofgauze is tied around the mandible and maxilla as a muzzle. Thetemplate device is placed evenly against the buccal mucosa, paral-lel to the lip margin, and triggered. Simultaneously, a stopwatchis started. Blood flow from the incision is blotted using circularfilter paper (Whatman No. 1; Fisher Scientific Co., Clifton, NJ)held directly below, but not touching the wound. The filter paperis changed every 15 s. The end point for each bleeding is deter-mined when the filter paper no longer develops a red crescent.


3.3.3. Evaluation The time from triggering the device until blood no longer appearson the filter paper is recorded as the bleeding time. The normalrange is 2–4 min.


The template bleeding time varies considerably between laborato-ries, as well as between species and strains. Therefore, it is impor-tant to perform the incisions and the blotting in an identical fash-ion. Prolonged bleeding times have been recognized in dogs withthrombocytopenia, von Willebrand’s disease and uremia, and indogs treated with aspirin, anticoagulants, and dextran (179, 180).Brassard and Meyers (181) reported that buccal mucosa bleed-ing time is sensitive to platelet adhesion and aggregation deficits.Generally, the effects of antithrombotic drugs in bleeding timemodels in animals do not exactly predict bleeding risks in clinicalsituations. However, the models allow comparison between drugswith different actions (174, 182).


The Simplate device can also be used to perform incisions in theshaved inner ear of rabbits, taking care to avoid major vessels.Normal bleeding time in anaesthetized rabbits is approximately100 s (77±4 s, n=20) (Just et al. unpublished data).

Klement et al. (180) described an ear bleeding model inanaesthetized rabbits. The shaved ear was immersed in a beakercontaining saline at 37◦C. Five full-thickness cuts were madewith a No. 11 Bard-Parker scalpel blade, avoiding major vessels,and the ear was immediately re-immersed in saline. At differ-ent times thereafter (5–30 min), aliquots of the saline solutionwere removed, red cells were sedimented and lysed, and cyanohe-moglobin was determined as a measure of blood loss. In thisstudy, hirudin produced more bleeding than standard heparin.

A cuticle bleeding time (toenail bleeding time) measurementin dogs has been described by Giles et al. (183). A guillotine-type toenail clipper is used to sever the apex of the nail cuticle. Aclean transection of the nail is made just into the quick to producea free flow of blood. The nail is left to bleed freely. The timeuntil bleeding stops is recorded as the bleeding time. Several nailscan be cut at one time to ensure appropriate technique. Normalbleeding time in this model ranges from 2 to 8 min.

4. GeneticModelsof Hemostasisand Thrombosis

4.1. Purposeand Rationale

Recent advances in genetics and molecular biology have providedtools that allow scientists to design genetically altered animals

80 Mousa

that are deficient in specific proteins involved in thrombosis andhemostasis (so-called knock-outs or nulls) (184, 185). These ani-mals have been extremely useful for identifying and validatingnovel targets for therapeutic intervention. That is, by examiningthe phenotype (e.g., spontaneous bleeding, platelet defects, pro-longed bleeding after surgical incision, etc.) of a specific knock-out strain, scientists can identify the role of the deleted pro-tein. If the phenotype is favorable (e.g., not lethal), pharmaco-logical agents can be designed to mimic the knockout. Morerecently, novel genetic-based medical approaches have also ben-efited greatly from the availability of these models, as discussedbelow. The following section briefly summarizes some of themajor findings in thrombosis and hemostasis using geneticallyaltered mice and concludes with an example of how these modelshave been used in the drug discovery process.

The majority of gene knockouts result in mice that developnormally, are born in the expected Mendelian ratios, and areviable, as defined by the ability to survive to adulthood. Althoughseemingly normal, these knock-out mice exhibit alterations inhemostatic regulation, especially when challenged. Deletion ofFVIII, FIX, von Willebrand factor and β3-integrin (186–189) allresult in mice that bleed upon surgical challenge, and despitesome minor differences in bleeding susceptibility, these mouseknock-out models mirror the human disease states quite well(hemophilia A, hemophilia B, von Willebrand disease, and Glanz-mann’s thrombasthenia, respectively). Deletion of some hemo-static factors results in fragile mice with severe deficiencies intheir ability to regulate blood loss. Prenatally, these mice appearto develop normally, but they are unable to survive the perina-tal period due to severe hemorrhaging, in most cases due to thetrauma of birth.

Genetic knockouts have also been useful in dissecting the roleof individual signaling proteins in platelet activation. Deletion ofβ3-integrin (188) or of Gαq (190) results in dramatic impairmentof agonist-induced platelet aggregation. Alteration of the proteincoding region in the β3-integrin carboxy tail, β3-DiY, at sites thatare thought to be phosphorylated upon platelet activation alsoresults in unstable platelet aggregation (191). Deletion of vari-ous receptors, such as thromboxane A2, P-selectin, P2Y1, andPAR-3, results in diminished responses to some agonists, whileother platelet responses are intact (192–195). Deletion of PAR-3,another thrombin receptor in mice, had little effect on hemosta-sis, indicating the presence of yet another thrombin receptor inplatelets, leading to the identification of PAR-4 (192).

Given that knockouts of prothrombotic factors yield micewith bleeding tendencies, it follows that deletion of factors in thefibrinolytic pathway results in increased thrombotic susceptibilityin mice. Plasminogen (196, 197), t-PA, urokinase-type plasmino-


gen activator (u-PA), and combined t-PA/u-PA double knock-out (198) result in mice with impaired fibrinolysis, susceptibilityto thrombosis, vascular occlusion, and tissue damage due to fib-rin deposition. Interestingly, due to fibrin formation in the heart,these mice might serve as good models of myocardial infarctionand heart failure caused by thrombosis (199). Intriguingly, micedeficient in PAI-1, the primary inhibitor of plasminogen activa-tor, exhibit no spontaneous bleeding and a greater resistance tovenous thrombosis due to a mild fibrinolytic state (200), whichsuggests that inhibition of PAI-1 might be a promising target fornovel antithrombotic agents.

In addition to their role in the regulation of hemostasis, sev-eral targeted genes are important in embryonic development. Forexample, deletion of tissue factor (201–203), tissue factor path-way inhibitor (TFPI) (204), or thrombomodulin (205) resultsin an embryonic lethal phenotype. These and other (206, 207)hemostatic factors also appear to contribute to vascular integrityin the developing embryo. These data suggest that initiation ofcoagulation and generation of thrombin is important at a criti-cal stage of embryonic development, yet other factors must con-tribute, since some of these embryos are able to progress andsurvive to birth.

Clearly, genetically altered mice have provided valuableinsight into the roles of specific hemostatic factors in physiologyand pathophysiology. The results of studies using knock-out micehave provided rationale and impetus for attacking certain targetspharmacologically. Knock-out mice have also provided excellentmodel systems for studying novel treatments for human dis-eases. For example, genetically altered animals have providedexceptional systems for the development of gene therapy forhemophilia. Specifically, deletion of FIX, generated by specificdeletions in the FIX gene and its promoter, results in mice witha phenotype that mimics the human phenotype of hemophiliaB (208). When these mice are treated by adenovirus-mediatedtransfer of human FIX, the bleeding diathesis is fully corrected(209). Similarly, selectively bred dogs that have a characteris-tic point mutation in the gene sequence encoding the catalyticdomain of FIX also have a severe hemophilia B that is pheno-typically similar to the human disease (210). Adeno-associatedvirus-mediated delivery of the canine FIX gene to these dogsintramuscularly resulted in measurable, therapeutic levels of FIXfor up to 17 months (211). Clinically relevant partial recoveryof whole blood clotting time and aPTT was also observed overthis prolonged period. These data provided support for the firststudy of adeno-associated virus-mediated FIX gene transfer inhumans (212). Preliminary results from clinical studies showedevidence of expression of FIX in three hemophilia patients andalso provided favorable safety data to substantiate the use of this

82 Mousa

therapeutic at higher doses. Although it is likely that there aredifferences between the human disease and animal models ofhemophilia (or other diseases), it is clear that these types of exper-iments can provide pharmacological, pharmacokinetic, and safetydata that will be extremely useful in designing and developingapproaches for safe clinical trials.

Gene therapy for patients with bleeding diatheses is moreadvanced than gene therapy for thrombotic indications. How-ever, promising preclinical data indicates that local overexpres-sion of thrombomodulin (213) or t-PA (214) inhibits throm-bus formation in a rabbit model of arterial thrombosis. Similarly,local gene transfer of TFPI prevented thrombus formation inballoon-injured porcine carotid arteries (215). These and otherstudies (216) suggest that novel gene therapy approaches willalso be effective for thrombotic indications, but these treatmentswill need to be carefully optimized in terms of pharmacokinetics,safety, and efficacy in laboratory animal studies prior to testing inhumans.

4.1.1. Knock-out Mice

4.1.1.1. Factor I(Fibrinogen)

Mice are normal in appearance at birth. Approximately 10% dieshortly after birth, and another 40% at 1–2 months after birthdue to bleeding and/or failure of pregnancy. Blood fails to clotor support platelet aggregation in vitro (217).

4.1.1.2. Factor II(Prothrombin)

Factor II deletion is a partial embryonic lethal mutation, with a50% rate of death between embryonic day (E) 9.5 and E11.5. Atleast 25% survive to term, but suffer from fatal hemorrhage a fewdays after birth. Factor II is important for maintaining vascularintegrity during development and post-natally (218, 219).

4.1.1.3. Factor V Half of the embryos die at E9–E10, possibly as a result of abnor-mal yolk sac vasculature. The remaining 50% progress normally toterm, but die from massive hemorrhage within 2 h of birth. Thisdefect results in a more severe phenotype in the mouse than inhuman (207, 218).

4.1.1.4. Factor VII Mice develop normally but suffer fatal perinatal bleeding (219).

4.1.1.5. Factor VIII Mice exhibit a mild phenotype as compared to severe hemophiliaA in humans. There is no spontaneous bleeding, illness or reducedactivity during the first year of life. Blood exhibits residual clottingactivity (aPTT) (220).

4.1.1.6. Factor IX Factor IX coagulant activity (aPTT) associated with wild-type,heterozygous, and homozygous null mice is as follows: +/+, 92%;+/−, 53%; −/−, <5%, respectively. Mice suffer from a bleed-ing disorder, which has been characterized as extensive bleeding


after clipping a portion of the tail, with bleeding to death if notcauterized (189, 221).

4.1.1.7. Factor X Partial embryonic lethal (1/3 of the mice died on E11.5 orE12.5), with fatal neonatal bleeding between perinatal day 5 (P5)and P20 (222).

4.1.1.8. Factor XI APTT is prolonged in homozygous null (−/−) mice (158–200 s)as compared to wild-type (+/+, 25–34 s) and heterozygous (+/−,40–61 s) mice. No factor XI activity; knockout does not resultin intrauterine death; homozygous null exhibit similar bleed-ing times as wild-type mice, with a tendency to prolongation(223).

4.1.1.9. TF (TissueFactor)

Abnormal circulation from yolk sac to embryo at approximatelyE8.5, leading to embryo wasting and death, most likely reflect-ing the role of TF in blood vessel development (201–203,224).

4.1.1.10. TFPI Lethal, with no survivors beyond the neonatal period. Approxi-mately 60% of mice die between E9.5 and E11.5, with signs ofyolk sac hemorrhage (204).

4.1.1.11. ThrombinReceptor (TR)

Approximately 50% of the mice die at E9–E10. Fifty percentsurvive and become normal adult mice at the gross level, with nobleeding diathesis. Null platelets strongly respond to thrombin,whereas null fibroblasts loose their ability to respond to thrombin,indicating the existence of a second TR (206, 225).

4.1.1.12.Thrombomodulin

The null mutation is an embryonic lethal, with embryosdying before the development of a functional cardiovascu-lar system. Mice die before E9.5 due to growth retardation.Heterozygous mice develop normally without thrombotic com-plications (199, 205, 226, 227).

4.1.1.13. Protein C Mice appear to develop normally at the macroscopic level, butexhibit obvious signs of bleeding and thrombosis. Mice do notsurvive beyond 24 h after delivery. Microvascular thrombosis inthe brain and necrosis in the liver is observed. Plasma clottablefibrinogen is undetectable, suggesting fibrinogen depletion andsecondary consumptive coagulopathy (228).

4.1.1.14. Plasminogen Mice exhibit severe spontaneous thrombosis, reduced ovula-tion and fertility, cachexia and short survival, severe glomeru-lonephritis, impaired skin healing, and reduced macrophage andkeratinocyte migration (196, 197).

4.1.1.15.Alpha2-antiplasmin

Mice exhibit normal fertility, viability, and development; nobleeding disorders; spontaneous lysis of injected clots, which isindicative of enhanced fibrinolytic potential; and a significant

84 Mousa

reduction of renal fibrin deposition after lipopolysaccharide (LPS)administration (229).

4.1.1.16. t-PA Mice exhibit extensive spontaneous fibrin deposition, severespontaneous thrombosis, impaired neointima formation, reducedovulation and fertility, cachexia and short survival, severeglomerulonephritis, and abnormal tissue remodeling (198, 199).

4.1.1.17. PAI-1 Mice exhibit reduced thrombotic incidence, no bleeding, accel-erated neointima formation, reduced lung inflammation andreduced atherosclerosis. Detailed studies of PAI-1 knock-out micehave been reported by Carmeliet et al. (200), Eitzman et al.(230), Erickson et al. (231), Kawasaki et al. (212), and Pinskyet al. (232).

4.1.1.18. Vitronectin Mice exhibit normal development, fertility, and survival; serum iscompletely deficient in “serum spreading factor” and PAI-1 bind-ing activity. Mice also exhibit delayed arterial and venous throm-bus formation (233, 234).

4.1.1.19. Urokinase,u-PA (Urinary-TypePlasminogen Activator)

Single u-PA knock-out mice are viable, fertile, and have a normallife span. Mice occasionally exhibit spontaneous fibrin depositionin normal and inflamed tissue and have a higher incidenceof endotoxin-induced thrombosis. Combined t-PA/u-PA-knock-out mice survive embryonic development, but exhibit retardedgrowth, reduced fertility, and shortened life span; spontaneousfibrin deposits are more extensive and occur in more organs(198, 235).

Transgenic mice carrying the u-PA gene linked to the albuminenhancer/promoter exhibit spontaneous intestinal and intra-abdominal bleeding that is directly related to transgene expres-sion in the liver and elevated plasma u-PA levels. Approximately50% of the transgenic mice die between 3 and 84 h after birthwith severe hypofibrinogenemia and loss of clotting function.

4.1.1.20. UPAR(Urinary-TypePlasminogen ActivatorReceptor)

Mice are phenotypically normal with attenuated thrombocytope-nia and mortality associated with severe malaria (236–239).

4.1.1.21. Gas 6 (GrowthArrest-Specific Gene 6Product)

Mice are viable, fertile, and appear normal; they do not sufferspontaneous bleeding or thrombosis and have normal tail bleed-ing times. Platelets fail to aggregate irreversibly to ADP, colla-gen, or U 46619. Arterial and venous thrombosis is inhibited andmice are protected from fatal thromboembolism after injection ofcollagen plus epinephrine (168).

4.1.1.22. GPIbα(Glycoprotein Ibα, Partof the GPIb-V–IXComplex)

Mice exhibit bleeding, thrombocytopenia and giant platelets(similar to human Bernard Soulier syndrome) (240).


4.1.1.23. GPV(Glycoprotein V, Part ofthe GPIb-V–IX Complex)

Mice exhibit increased thrombin responsiveness; GPV nullplatelets are normal in size; mice have normal amounts of GPIb-IX and functional von Willebrand factor (vWF) binding. Nullplatelets are hyper-responsive to thrombin and have an increasedaggregation response and shorter bleeding time, reflecting theactivity of GPV as a negative modulator of platelet function (241).

4.1.1.24. GPIIb (IntegrinAlpha IIb, GlycoproteinIIb, Part of the GPIIb–IIIaComplex)

Mice have a bleeding disorder similar to Glanzmann thrombas-thenia in man. Null platelets fail to bind fibrinogen, aggregateand retract a fibrinogen clot; platelet granules do not containfibrinogen (242).

4.1.1.25. GPIIIa (IntegrinBeta3, Glycoprotein IIIa,Part of the GPIIb–IIIaComplex)

Mice are viable and fertile, but have increased fetal mortality. Miceexhibit features of Glanzmann thrombasthenia in man, i.e., defec-tive platelet aggregation and clot retraction, spontaneous bleed-ing, prolonged bleeding times, dysfunctional osteoclasts, and thedevelopment of osteosclerosis with age (188, 243).

4.1.1.26. GPIIa(Glycoprotein IIa,Integrin β1, Part of theGPIa–IIa Complex)

Integrin β1 null platelets from conditional knock-out micedevelop normally, and platelet count is normal; collagen-inducedplatelet aggregation is delayed but otherwise normal; thetyrosine phosphorylation pattern is normal but phosphorylationis delayed. The bleeding time in the bone marrow of chimericmice is normal, and there are no major in vivo defects (244).

4.1.1.27. vWF FVIII levels are strongly reduced due to defective protectionby vWF. Mice exhibit highly prolonged bleeding times, hemor-rhaging, and spontaneous bleeding. vWF knockout mice havebeen useful for investigating the role of vWF. Mice exhibitdelayed platelet adhesion in ferric chloride-induced arteriolarinjury (187, 245).

4.1.1.28. ThromboxaneA2 Receptor (TXA2R)

Mice exhibit a mild bleeding disorder and altered vascularresponses to TXA2 and arachidonic acid (195).

4.1.1.29. ProstacyclinReceptor (PGI2R)

Mice are viable, fertile, and normotensive, with increased sus-ceptibility to thrombosis and reduced inflammatory and painresponses (246).

4.1.1.30. PECAM(Platelet: Endothelial CellAdhesion Molecule)

Mice exhibit normal platelet aggregation; Duncan et al. (247)and Mahooti et al. (248) described prolonged bleeding times inPECAM knockout mice.

4.1.1.31. Pallid (Pa) The pallid mouse, one of 13 hypo-pigment mouse mutants witha storage pool deficiency, is a model of human Herman sky Pud-lak syndrome (the beige mouse is a model of Chediak Higashisyndrome). Pallid mice exhibit prolonged bleeding times, pig-ment dilution, elevated kidney lysosomal enzyme, serum α1 antit-rypsin deficiency and abnormal otolith formation. The genedefective in pallid mice encodes the highly charged 172-amino

86 Mousa

acid protein pallidin, which interacts with syntaxin 13, a proteinthat mediates vesicle docking and fusion (249).

4.1.1.32. Gα(q) (GuanylNucleotide BindingProtein Gαq)

Blood from knock-out mice exhibits defective aggregation inresponse to ADP, TXA2, thrombin, and collagen; shape change isnormal (190, 250).

4.1.1.33. Gz (A Memberof the Gi Familyof G Proteins)

Mice exhibit impaired platelet aggregation in response toepinephrine, resistance to fatal thromboembolism, and exagger-ated responses to cocaine. Morphine and antidepressant drugshave a reduced effect in knock-out mice (169).

4.1.1.34. PhospholipaseCγ

Mice are viable and fertile, with decreased numbers of matureB cells, defective B cell and mast cell function, and defective Fcγreceptor signaling, resulting in a loss of collagen-induced plateletaggregation (251).

4.1.1.35. CD39(Vascular ATPDiphosphohydrolase)

Mice are viable and fertile, with prolonged bleeding timesbut minimally perturbed coagulation parameters. Mice exhibitreduced platelet interaction with injured mesenteric vasculature invivo; failure of platelets to aggregate in response to standard ago-nists in vitro is associated with purinergic P2Y1 receptor desensi-tization and fibrin deposition at multiple organ sites (252).

4.1.1.36. Protein Kinase,cGMP-Dependent,Type 1

Mice are viable and fertile, unresponsive to cGMP and NO,and defective in vasodilator-stimulated phosphoprotein (VASP)phosphorylation. Mice exhibit increased adhesion and aggre-gation of platelets in vivo in ischemic/re-perfused mesentericmicrocirculation and no compensation by the cAMP kinasesystem (253).

4.1.1.37. VASP Mice are viable and fertile and exhibit mild platelet dysfunctionwith megakaryocyte hyperplasia, increased collagen/thrombinactivation, and impaired cyclic nucleotide-mediated inhibition ofplatelet activation (254, 255).

4.1.1.38. Arachidonate12-Lipoxygenase(P-12LO)

Platelets exhibit a selective hypersensitivity to ADP, manifestedas a marked increase in slope and percent aggregation in ex vivoassays, and increased mortality in an ADP-induced mouse modelof thromboembolism (256, 257).

4.1.1.39. Arachidonate5-Lipoxygenase (P-5LO)

Mice develop normally and are healthy. They exhibit no differ-ences in reaction to endotoxin shock as compared to wild-typemice; however, they are resistant to the lethal effects of shockinduced by PAF. Inflammation induced by arachidonic acid inthese mice is markedly reduced (256).

4.1.1.40.Thrombopoietin

Thrombopoietin-null and thrombopoietin receptor (c-Mpl)-nullmice exhibit a 90% reduction in megakaryocyte and platelet levels.


However, despite these low platelet levels, mice do not exhibitexcessive bleeding. Platelets that are present are morphologicallynormal and functional, indicating that in vivo, thrombopoietin isrequired for control of megakaryocyte and platelet number, butnot maturation (258).

4.1.1.41.Thrombospondin-1

Mice exhibit normal thrombin-induced platelet aggrega-tion and higher numbers of circulating white blood cells.

Table 2.3Genetic models of thrombosis and hemostasis

Gene target Viable Embryonic development/survival References

Coagulation

Protein C No Normal/perinatal death (228)Fibrinogen Yes Normal/perinatal death (217)

Fibrinogen-QAGVD Yes Normal (217)FV No Partial embryonic loss/perinatal death (207)

FVII Yes Normal/perinatal death (219)FVIII Yes Normal (186)

FIX Yes Normal (189)FXI Yes Normal (223)

Tissue factor No Lethal (201, 203)TFPI No Lethal (204)

vWF Yes Normal (187)Prothrombin No Partial embryonic loss/perinatal death (277, 278)

Fibrinolyticu-PA and t-PA Yes Normal/growth retardation (198)

uPAR Yes Normal (237, 238)Plasminogen Yes Normal/growth retardation (196, 197)

PAI Yes Normal (200)Thrombomodulin No Lethal (205)

PlateletB3 Yes Normal/partial embryonic loss (188)

B3-DiYF Yes Normal (191)P-Selectin Yes Normal (194)

PAR-1 Yes Normal (206)PAR-3 Yes Normal (192)

Gαq Yes Normal/perinatal death (190)TXA2 receptor Yes Normal (195)

P2Y1 Yes Normal (193)

88 Mousa

Thrombospondin-1 has been implicated in normal lung home-ostasis (258).

Mouse knock-out models of virtually all of the known hemo-static factors have been reported, as shown in Table 2.3.

5. Critical Issuesin Experimentalmodels

5.1. The Useof Positive Controls

Clearly, there are many antithrombotic agents that can be usedto compare and contrast the antithrombotic efficacy and safetyof novel agents. The classic antithrombotic agents are heparin,warfarin, and aspirin. However, new, more selective agents suchas hirudin, LMWHs, and clopidogrel are commercially availablethat will either replace or augment these older treatments. Novelantithrombotic agents should certainly be demanded to demon-strate better efficacy than currently available therapy in animalmodels of thrombosis. This should be demonstrated by perform-ing dose–response experiments that include maximally effectivedoses of each compound in the model. At the maximally effec-tive dose, parameters such as aPTT, PT, template bleeding time,or other, more sensitive measurements of systemic hypocoagula-bility or bleeding should be compared. A good example of thisapproach is a study by Schumacher et al. (259), who comparedthe antithrombotic efficacy of argatroban and deltaparin in arte-rial and venous models of thrombosis. Consideration of potencyand safety compared to other agents should be taken into accountwhen advancing a drug through the testing funnel.

The early in vivo evaluation of compounds that demonstrateacceptable in vitro potency and selectivity requires evaluationof each compound alone in order to demonstrate antithrom-botic efficacy. The antithrombotic landscape is becoming com-plicated by so many agents from which to choose that it willbecome increasingly difficult to design preclinical experimentsthat mimic the clinical setting in which poly-antithrombotictherapy is required for optimal efficacy and safety. Consequently,secondary and tertiary preclinical experiments will need to becarefully designed in order to answer these specific, importantquestions.

5.2. Evaluationof Bleeding Tendency

Although the clinical relevance of animal models of thrombo-sis has been well-established in terms of efficacy, the preclini-cal tests for evaluating safety, i.e., bleeding tendency, have notbeen as predictable. The difficulty in predicting major bleeding,such as intracranial hemorrhage, resulting from antithrombotic


or thrombolytic therapy stems from the complexity and lackof understanding of the mechanisms involved in this disorder.Predictors of anticoagulant-related intracranial hemorrhage areadvanced age, hypertension, intensity and duration of treatment,head trauma, and prior neurologic disease (260, 261). These riskfactors are clearly difficult, if not impossible, to simulate in labora-tory animals. Consequently, more general tests of anticoagulationand primary hemostasis have been employed.

Coagulation assays provide an index of the systemic hypoco-agulability of the blood after administration of antithromboticagents; however, as indicated earlier, the sensitivity and speci-ficity of these assays vary from compound-to-compound, so theseassays do not provide a consistent safety measure across allmechanisms of inhibition. Consequently, many laboratories haveattempted to develop procedures that provide an indication ofbleeding risk by evaluating primary hemostasis after generatingcontrolled incisions in anesthetized animals. Some of the testsused in evaluating FXa inhibitors include template bleeding time,tail transection bleeding time, cuticle bleeding time, and evalu-ation of clinical parameters such as hemoglobin and hematocrit.Unfortunately, template bleeding tests, even when performed inhumans, have not been good predictors of major bleeding eventsin clinical trials (262–264). However, these tests have been able todemonstrate relative advantages of certain mechanisms and agentsover others. For example, hirudin, a direct thrombin inhibitor,appears to have a narrow therapeutic window when used as anadjunct to thrombolysis in clinical trials, producing unacceptablemajor bleeding when administered at 0.6 mg/kg i.v. bolus, plus0.2 mg/kg/h (265, 266). When the dose of hirudin was adjustedto avoid major bleeding (0.1 mg/kg and 0.1 mg/kg/h), no sig-nificant therapeutic advantage over heparin was observed. If therelative improvement in the ratio between efficacy and bleedingobserved preclinically with FXa inhibitors compared to throm-bin inhibitors such as hirudin is supported in future clinical trials,this will establish an important safety advantage for FXa inhibitorsand provide valuable information for evaluating the safety of newantithrombotic agents in preclinical experiments.

5.3. Selectionof Models Based onSpecies-DependentPharmacology/Physiology

As alluded to earlier, species selection for animal models ofdisease is often limited by the unique physiology of a partic-ular disease target in different species or by the species speci-ficity of the pharmacological agent for the target. For example,it was discovered relatively early in the development of plateletGPIIb/IIIa antagonists that these compounds were of limiteduse in rats (267) and that there was a dramatic species-dependentvariation in the response of platelets to GPIIb/IIIa antagonists(268–270). This discovery led to the widespread use of larger ani-mals (particularly dogs, whose platelet response to GPIIb/IIIa

90 Mousa

antagonists resembles humans) in the evaluation of GPIIb/IIIaantagonists. Of course the larger animals required more com-pound for evaluation, which created a resource problem formedicinal chemists. This was especially problematic for compa-nies that generated compounds by combinatorial parallel syn-thetic chemistry in which many compounds can be made, butusually in very small quantities. However, some pharmacologistsdevised clever experiments that partially overcame this problem.Cook et al. (271) administered a GPIIb/IIIa antagonist orallyand intravenously to rats and then mixed platelet-rich plasmafrom the treated rats with platelet-rich plasma from untreateddogs. The mixture was then evaluated in an agonist-inducedplatelet aggregation assay and the resulting inhibition of canineplatelet aggregation (rat platelets were relatively unresponsive tothis GPIIb/IIIa antagonist) was due to the drug present in theplasma obtained from the rat. Using this method, only a smallamount of drug is required to determine the relative bioavailabil-ity in rats. However, the animal models chosen for efficacy in thatreport (guinea pigs and dogs) were selected based on their favor-able platelet response to the GPIIb/IIIa antagonist.

Similarly for inhibitors of FXa, there are significant varia-tions in the activity of certain compounds against FXa purifiedfrom plasma of different species and in plasma-based clottingassays using plasma from different species. DX-9065 is muchmore potent against human FXa (Ki=78 nM) than against rabbit(Ki=102 nM) and rat (Ki=1980 nM) FXa. Likewise, in the PTassay, DX-9065a was very potent in human plasma (concentra-tion required to double PT, PT×2, was 0.52 μM) and in squirrelmonkey plasma (PT×2 = 0.46 μM), but was much less potentin rabbit, dog, and rat plasma (PT×2 = 1.5, 6.5, and 22.2 μM,respectively). Other FXa inhibitors have also demonstrated thesespecies-dependent differences in activity (272–274). Regardless,the investigator must be aware of these differences so that appro-priate human doses can be extrapolated from the laboratory ani-mal studies.

Although in many cases the exact mechanism for the species-dependent differences in response to certain therapeutic agentsremains unclear, these differences must be examined to determinethe appropriate species to be used for preclinical pharmacologicalevaluation of each agent. This evaluation can routinely be per-formed by in vitro coagulation or platelet aggregation tests priorto evaluation in animal models.

5.4. Selectionof Models Basedon Pharmacokinetics

Much debate surrounds the issue as to which species most resem-bles humans in terms of gastrointestinal absorption, clearance,and metabolism of therapeutic agents. Differences in gastroin-testinal anatomy, physiology, and biochemistry between humansand commonly used laboratory animals suggest that no single


Tabl

e2.

4An

imal

mod

els

ofth

rom

bosi

san

dth

eirc

linic

alco

rrel

ates

Com

poun

dPr

eclin

ical

anim

alm

odel

Prec

linic

alre

sults

Refe

renc

eCl

inic

alin

dica

tion

Clin

ical

resu

ltRe

fere

nce

Rec

ombi

nant

t-PA

(Act

ivas

e)

Rab

bit

pulm

onar

yar

tery

thro

mbo

sis

Lys

isof

pref

orm

edpu

lmon

ary

thro

mbu

s(2

79)

Acu

tem

yoca

rdia

lin

farc

tion–

thro

mbo

lysi

s

Impr

oved

reca

naliz

atio

n(2

80)

Abc

ixim

ab(R

eoPr

o)C

anin

eco

rona

rycy

clic

flow

redu

ctio

n(4

4)Si

gnifi

cant

inhi

bitio

nof

plat

elet

-dep

ende

ntth

rom

bosi

s

(39)

Hig

h-ri

skco

rona

ryan

giop

last

yR

educ

tion

inde

ath,

myo

card

iali

nfar

ctio

n,re

frac

tory

isch

emia

,or

unpl

anne

dre

vasc

ular

izat

ion

(281

)

Tir

ofiba

n(A

ggre

stat

)C

anin

eco

rona

rycy

clic

flow

redu

ctio

n(4

4)Si

gnifi

cant

inhi

bitio

nof

plat

elet

-dep

ende

ntth

rom

bosi

s

(282

)U

nsta

ble

angi

naR

educ

tion

inde

ath,

myo

card

iali

nfar

ctio

n,re

frac

tory

isch

emia

(283

)

Ept

ifibi

tide

(Int

egri

lin)

TPA

-ind

uced

coro

nary

thro

mbo

lysi

sSi

gnifi

cant

impr

ovem

ent

inly

sis

ofoc

clus

ive

thro

mbu

s

(81)

Acu

tem

yoca

rdia

lin

farc

tion–

thro

mbo

lysi

sw

itht-

PA

Impr

ovem

ent

inin

cide

nce

and

spee

dof

repe

rfus

ion

(284

)

Eno

xapa

rin

(Lov

enox

)C

anin

eco

rona

rycy

clic

flow

redu

ctio

n(4

4)Si

gnifi

cant

inhi

bitio

nof

plat

elet

-dep

ende

ntth

rom

bosi

s

(50)

Uns

tabl

ean

gina

Sign

ifica

ntde

crea

sein

deat

h,m

yoca

rdia

lin

farc

tion,

and

need

for

reva

scul

ariz

atio

nat

30da

ys

(285

)

Hir

udin

(Refl

udan

)R

abbi

tju

gula

rve

inth

rom

bus

grow

thIn

hibi

tion

ofth

rom

bus

grow

thco

mpa

red

tost

anda

rdhe

pari

n

(286

)D

eep

vein

thro

mbo

sis

afte

rto

talh

ipre

plac

emen

t

Sign

ifica

ntly

decr

ease

dra

teof

DV

T(2

87)

Arg

atro

ban

Can

ine

coro

nary

arte

ryel

ectr

olyt

icin

jury

(TPA

-ind

uced

thro

mbo

lysi

s)

Acc

eler

ated

repe

rfus

ion

and

prev

ente

dre

-occ

lusi

on

(288

)U

nsta

ble

angi

naN

oep

isod

esof

MI

duri

ngdr

ugin

fusi

on(2

89)

92 Mousa

animal can precisely mimic the gastrointestinal characteristics ofhumans (275). Due to resource issues (mainly compound avail-ability) and animal care and use considerations, small rodents suchas rats are usually considered for primary in vivo evaluation ofpharmacokinetics for novel agents. However, there is great reser-vation about moving a compound into clinical trials based on oralbioavailability data derived from rat experiments alone. Usually,larger animals such as dogs or non-human primates, which havesimilar gastrointestinal morphology compared to humans, are thenext step in the evaluation of pharmacokinetics of new agents.The pharmacokinetic characteristics of the FXa inhibitor YM-60828 have been studied extensively in a variety of laboratory ani-mals. YM-60828 demonstrated species-dependent pharmacoki-netics, with oral bioavailability estimates of approximately 4, 33,7, and 20% in rats, guinea pigs, beagle dogs, and squirrel mon-keys, respectively. Although these results suggest that YM-60828has somewhat limited bioavailability, evaluating the pharmacoki-netic profile of novel agents in a number of species (276) is a well-established approach used to aid in identifying compounds foradvancement to human testing. That is, acceptable bioavailabilityin a number of species suggests that a compound will be bioavail-able in humans. Which of the laboratory species adequately rep-resents the bioavailability of a specific compound in humans canonly be determined after appropriate pharmacokinetic evalua-tion in humans. Nevertheless, preclinical pharmacokinetic dataare important in selecting the appropriate animal model for test-ing the antithrombotic efficacy of compounds because the ulti-mate proof-of-concept experiment is to demonstrate efficacy bythe intended route of administration.

6. ClinicalRelevanceof Data Derivedfrom ExperimentalModels

Animal models of thrombosis have played a crucial role in thediscovery and development of a number of compounds that arenow successfully being used for the treatment and preventionof thrombotic diseases. Influential preclinical results using novelantithrombotics in a variety of laboratory animal experiments arelisted in Table 2.4, along with the early clinical trials and resultsfor each compound. This table intentionally omits many com-pounds that were tested in animal models of thrombosis, butfailed to be successful in clinical trials or, for other reasons, didnot become approved drugs. However, these negative outcomeswould not have been predicted by animal models of thrombo-sis because the failures were generally due to other shortcom-ings of the drugs (e.g., toxicity, narrow therapeutic window, or


undesirable pharmacokinetics or pharmacodynamics) which arenot always clearly presented in the scientific literature due to pro-prietary restrictions in this highly competitive field.

Nonetheless, it is clear that animal models have supplied valu-able information for investigators responsible for evaluating thesedrugs in humans, providing pharmacodynamic, pharmacokinetic,and safety data that can be used to design safe and efficient clinicaltrials. The reader is referred to a number of detailed reports onthe applications of animal models (188, 190–195, 197–213, 215,216, 237).

References

1. Didisheim, P. (1972) Animal models usefulin the study of thrombosis and antithrom-botic agents Prog Hemost Thromb 1,165–97.

2. Leadley, R.J., Jr., Chi, L., Rebello, S.S., andGagnon, A. (2000) Contribution of in vivomodels of thrombosis to the discovery anddevelopment of novel antithrombotic agentsJ Pharmacol Toxicol Methods 43, 101–16.

3. Wessler, S., Reimer, S.M., and Sheps, M.C.(1959) Biologic assay of a thrombosis-inducing activity in human serum J Appl Phys-iol 14, 943–6.

4. Meuleman, D.G., Hobbelen, P.M., VanDinther, T.G., Vogel, G.M., Van Boeckel,C.A., and Moelker, H.C. (1991) Antifac-tor Xa activity and antithrombotic activity inrats of structural analogues of the minimumantithrombin III binding sequence: discov-ery of compounds with a longer durationof action than of the natural pentasaccharideSemin Thromb Hemost 17 (Suppl 1), 112–7.

5. Carrie, D., Caranobe, C., Saivin, S., Houin,G., Petitou, M., Lormeau, J.C., VanBoeckel, C., Meuleman, D., and Boneu,B. (1994) Pharmacokinetic and antithrom-botic properties of two pentasaccharideswith high affinity to antithrombin III in therabbit: comparison with CY216 Blood 84,2571–7.

6. Walenga, J.M., Petitou, M., Lormeau, J.C.,Samama, M., Fareed, J., and Choay, J.(1987) Antithrombotic activity of a syn-thetic heparin pentasaccharide in a rab-bit stasis thrombosis model using differentthrombogenic challenges Thromb Res 46,187–98.

7. Vlasuk, G.P., Ramjit, D., Fujita, T.,Dunwiddie, C.T., Nutt, E.M., Smith, D.E.,and Shebuski, R.J. (1991) Comparison ofthe in vivo anticoagulant properties of stan-dard heparin and the highly selective factorXa inhibitors antistasin and tick anticoagulant

peptide (TAP) in a rabbit model of venousthrombosis Thromb Haemost 65, 257–62.

8. Millet, J., Theveniaux, J., and Brown, N.L.(1994) The venous antithrombotic profile ofnaroparcil in the rabbit Thromb Haemost 72,874–9.

9. Callas, D.D., Bacher, P., and Fareed, J.(1995) Studies on the thrombogenic effectsof recombinant tissue factor. In vivo versusex vivo findings Semin Thromb Hemost 21,166–76.

10. Yeghiazarians, Y., Braunstein, J.B., Askari, A.,and Stone, P.H. (2000) Unstable angina pec-toris N Engl J Med 342, 101–14.

11. Anderson, H., and Willerson, J. Experi-mental models of thrombosis. In: Loscalzo,J., and Schafer, A., eds. Thrombosis andHemorrhage. Boston: Blackwell Scientific;1994:385–93.

12. Badimon, L. (1997) Models to study throm-botic disorders Thromb Haemost 78, 667–71.

13. Bush, L.R., and Shebuski, R.J. (1990) In vivomodels of arterial thrombosis and thrombol-ysis FASEB J 4, 3087–98.

14. Chi, L., Rebello, S., and Lucchesi, B.R. Invivo models of thrombosis. In: Uprichard,A., and Gallagher, K., eds. Anti-thrombotics.Heidelberg: Springer; 1999:101–27.

15. Gold, H.K., Yasuda, T., Jang, I.K., Guerrero,J.L., Fallon, J.T., Leinbach, R.C., andCollen, D. (1991) Animal models for arterialthrombolysis and prevention of reocclusion.Erythrocyte-rich versus platelet-rich throm-bus Circulation 83, IV26–40.

16. Bocan, T.M. (1998) Animal models ofatherosclerosis and interpretation of drugintervention studies Curr Pharm Des 4,37–52.

17. Johnson, G.J., Griggs, T.R., and Badimon,L. (1999) The utility of animal modelsin the preclinical study of interventions toprevent human coronary artery restenosis:analysis and recommendations. On behalf

94 Mousa

of the Subcommittee on Animal, Cellularand Molecular Models of Thrombosis andHaemostasis of the Scientific and Standard-ization Committee of the International Soci-ety on Thrombosis and Haemostasis ThrombHaemost 81, 835–43.

18. Hunter, A.J., Green, A.R., and Cross, A.J.(1995) Animal models of acute ischaemicstroke: can they predict clinically successfulneuroprotective drugs? Trends Pharmacol Sci16, 123–8.

19. Folts, J.D., Crowell, E.B., Jr., and Rowe,G.G. (1976) Platelet aggregation in partiallyobstructed vessels and its elimination withaspirin Circulation 54, 365–70.

20. Golino, P., Ragni, M., Cirillo, P., D’Andrea,D., Scognamiglio, A., Ravera, A., Buono,C., Ezban, M., Corcione, N., Vigorito,F., Condorelli, M., and Chiariello, M.(1998) Antithrombotic effects of recombi-nant human, active site-blocked factor VIIain a rabbit model of recurrent arterial throm-bosis Circ Res 82,39–46.

21. Aiken, J.W., Gorman, R.R., and Shebuski,R.J. (1979) Prevention of blockage of par-tially obstructed coronary arteries withprostacyclin correlates with inhibition ofplatelet aggregation Prostaglandins 17,483–94.

22. Aiken, J.W., Shebuski, R.J., Miller, O.V., andGorman, R.R. (1981) Endogenous prostacy-clin contributes to the efficacy of a throm-boxane synthetase inhibitor for preventingcoronary artery thrombosis J Pharmacol ExpTherap 219, 299–308.

23. Bush, L.R., Campbell, W.B., Buja, L.M.,Tilton, G.D., and Willerson, J.T. (1984)Effects of the selective thromboxane syn-thetase inhibitor dazoxiben on varia-tions in cyclic blood flow in stenosedcanine coronary arteries Circulation 69,1161–70.

24. Folts, J.D., Gallagher, K., and Rowe,G.G. (1982) Blood flow reductions instenosed canine coronary arteries: vasospasmor platelet aggregation? Circulation 65,248–55.

25. Shebuski, R.J., Ramjit, D.R., Bencen, G.H.,and Polokoff, M.A. (1989) Characteri-zation and platelet inhibitory activity ofbitistatin, a potent arginine-glycine-asparticacid-containing peptide from the venom ofthe viper Bitis arietans J Biol Chem 264,21550–6.

26. Shebuski, R.J., Ramjit, D.R., Sitko, G.R.,Lumma, P.K., and Garsky, V.M. (1990) Pre-vention of canine coronary artery throm-bosis with echistatin, a potent inhibitor ofplatelet aggregation from the venom of the

viper, Echis carinatus Thromb Haemost 64,576–81.

27. Uchida, Y., Yoshimoto, N., and Murao, S.(1975) Cyclic fluctuations in coronary bloodpressure and flow induced by coronary arteryconstriction Jpn Heart J 16, 454–64.

28. Bush, L.R., and Patrick, D.L. (1986) Therole of the endothelium in arterial thrombo-sis and the influence of antithrombotic ther-apy Drug Devel Res 7, 319–40.

29. Golino, P., Ashton, J.H., Buja, L.M.,Rosolowsky, M., Taylor, A.L., McNatt, J.,Campbell, W.B., and Willerson, J.T. (1989)Local platelet activation causes vasoconstric-tion of large epicardial canine coronary arter-ies in vivo. Thromboxane A2 and sero-tonin are possible mediators Circulation 79,154–66.

30. Bush, L.R., Campbell, W.B., Kern, K.,Tilton, G.D., Apprill, P., Ashton, J., Schmitz,J., Buja, L.M., and Willerson, J.T. (1984)The effects of alpha 2-adrenergic and sero-tonergic receptor antagonists on cyclic bloodflow alterations in stenosed canine coronaryarteries Circ Res 55, 642–52.

31. Bennett, J.S., and Vilaire, G. (1979) Expo-sure of platelet fibrinogen receptors byADP and epinephrine J Clin Invest 64,1393–401.

32. Coller, B.S., and Scudder, L.E. (1985) Inhi-bition of dog platelet function by in vivoinfusion of F(ab′)2 fragments of a mono-clonal antibody to the platelet glycoproteinIIb/IIIa receptor Blood 66, 1456–9.

33. Shebuski, R.J., Berry, D.E., Bennett, D.B.,Romoff, T., Storer, B.L., Ali, F., andSamanen, J. (1989) Demonstration ofAc-Arg-Gly-Asp-Ser-NH2 as an antiaggre-gatory agent in the dog by intracoronaryadministration Thromb Haemost 61, 183–8.

34. Eidt, J.F., Allison, P., Noble, S., Ashton,J., Golino, P., McNatt, J., Buja, L.M.,and Willerson, J.T. (1989) Thrombin is animportant mediator of platelet aggregationin stenosed canine coronary arteries withendothelial injury J Clin Invest 84, 18–27.

35. Smith, J., and Shebuski, R. (1988) Lack of aninhibitory effect of tissue plasminogen acti-vator (t-PA) on platelet aggregation in vivoFASEB J J2, A1164.

36. Al-Wathiqui, M.H., Hartman, J.C., Brooks,H.L., Gross, G.J., and Warltier, D.C. (1988)Cyclical carotid artery flow reduction in con-scious dogs: effect of a new thrombox-ane receptor antagonist Am Heart J 116,1482–7.

37. Gallagher, K.P., Folts, J.D., and Rowe,G.G. (1978) Comparison of coronary arteri-ograms with direct measurements of stenosed


coronary arteries in dogs Am Heart J 95,338–47.

38. Schumacher, W.A., Heran, C.L.,Goldenberg, H.J., Harris, D.N., andOgletree, M.L. (1989) Magnitude of throm-boxane receptor antagonism necessary forantithrombotic activity in monkeys Am JPhysiol 256, H726–34.

39. Coller, B.S., Folts, J.D., Scudder, L.E., andSmith, S.R. (1986) Antithrombotic effect ofa monoclonal antibody to the platelet glyco-protein IIb/IIIa receptor in an experimentalanimal model Blood 68, 783–6.

40. Eidt, J.F., Ashton, J., Golino, P., McNatt,J., Buja, L.M., and Willerson, J.T. (1989)Treadmill exercise promotes cyclic alterationsin coronary blood flow in dogs with coronaryartery stenoses and endothelial injury J ClinInvest 84, 517–27.

41. O’Neill, W., Topol, E., and Pitt, B. (1988)Coronary thrombolysis for evolving myocar-dial infarction Prog Cardiovasc Dis 30,465–83.

42. Folts, J., and Rowe, G. (1974) Cyclicalreductions in coronary blood flow in coro-nary arteries with fixed partial obstructionand their inhibition with aspirin Fed Proc 33.

43. Folts, J.D., and Rowe, G.G. (1988)Epinephrine potentiation of in vivo stimulireverses aspirin inhibition of platelet throm-bus formation in stenosed canine coronaryarteries Thromb Res 50, 507–16.

44. Folts, J. (1991) An in vivo model of exper-imental arterial stenosis, intimal damage,and periodic thrombosis Circulation 83,IV3–14.

45. Just, M., Tripier, D., and Seiffge, D.(1991) Antithrombotic effects of recom-binant hirudin in different animal modelsHaemostasis 21(Suppl 1), 80–7.

46. Apprill, P., Schmitz, J.M., Campbell, W.B.,Tilton, G., Ashton, J., Raheja, S., Buja,L.M., and Willerson, J.T. (1985) Cyclicblood flow variations induced by platelet-activating factor in stenosed canine coronaryarteries despite inhibition of thrombox-ane synthetase, serotonin receptors, andalpha-adrenergic receptors Circulation 72,397–405.

47. Romson, J.L., Hook, B.G., and Lucchesi,B.R. (1983) Potentiation of the antithrom-botic effect of prostacyclin by simultaneousadministration of aminophylline in a caninemodel of coronary artery thrombosis J Phar-macol Exp Therap 227, 288–94.

48. Just, M., and Schonafinger, K. (1991)Antithrombotic properties of a novel syd-nonimine derivative J Cardiovasc Pharmacol17(Suppl 3), S121–6.

49. Mousa, S.A., DeGrado, W.F., Mu, D.X.,Kapil, R.P., Lucchesi, B.R., and Reilly, T.M.(1996) Oral antiplatelet, antithromboticefficacy of DMP 728, a novel plateletGPIIb/IIIa antagonist Circulation 93,537–43.

50. Leadley, R.J., Jr., Kasiewski, C.J., Bostwick,J.S., Bentley, R., Dunwiddie, C.T., andPerrone, M.H. (1998) Inhibition of repet-itive thrombus formation in the stenosedcanine coronary artery by enoxaparin, butnot by unfractionated heparin ArteriosclThromb Vasc Biol 18, 908–14.

51. Rubsamen, K., and Kirchengast, M. (1998)Thrombin inhibition and intracoronarythrombus formation: effect of polyethy-lene glycol-coupled hirudin in the stenosed,locally injured canine coronary artery CoronArtery Dis 9, 35–42.

52. Mehta, J.L., Chen, L., Nichols, W.W.,Mattsson, C., Gustafsson, D., and Saldeen,T.G. (1998) Melagatran, an oral active-siteinhibitor of thrombin, prevents or delaysformation of electrically induced occlusivethrombus in the canine coronary arteryJ Cardiovasc Pharmacol 31, 345–51.

53. Ikeda, H., Ueyama, T., Murohara, T.,Yasukawa, H., Haramaki, N., Eguchi, H.,Katoh, A., Takajo, Y., Onitsuka, I., Ueno, T.,Tojo, S.J., and Imaizumi, T. (1999) Adhe-sive interaction between P-selectin and sialylLewis(x) plays an important role in recurrentcoronary arterial thrombosis in dogs Arte-rioscl Thromb Vasc Biol 19, 1083–90.

54. Jackson, C.V., Bailey, B.D., and Shetler,T.J. (2000) Pharmacological profile ofrecombinant, human activated protein C(LY203638) in a canine model of coronaryartery thrombosis J Pharmacol Exp Therap295, 967–71.

55. Romson, J.L., Haack, D.W., and Lucch-esi, B.R. (1980) Electrical induction ofcoronary artery thrombosis in the ambula-tory canine: a model for in vivo evaluationof anti-thrombotic agents Thromb Res 17,841–53.

56. Benedict, C.R., Mathew, B., Rex, K.A.,Cartwright, J., Jr., and Sordahl, L.A. (1986)Correlation of plasma serotonin changes withplatelet aggregation in an in vivo dog modelof spontaneous occlusive coronary thrombusformation Circ Res 58, 58–67.

57. Warltier, D.C., Lamping, K.A., Pelc, L.R.,and Gross, G.J. (1987) A canine model ofthrombin-induced coronary artery thrombo-sis: effects of intracoronary streptokinase onregional myocardial blood flow, contractilefunction, and infarct size J Pharmacol Meth-ods 18, 305–18.

96 Mousa

58. Coller, B.S., Folts, J.D., Smith, S.R.,Scudder, L.E., and Jordan, R. (1989) Abo-lition of in vivo platelet thrombus formationin primates with monoclonal antibodies tothe platelet GPIIb/IIIa receptor. Correlationwith bleeding time, platelet aggregation, andblockade of GPIIb/IIIa receptors Circula-tion 80, 1766–74.

59. Harbauer, G., Hiller, W., and Hellstern, P.An experimental model of venous thrombo-sis in the rabbit; test of its usefulness withlow-dose heparin. In: L, K., ed. Chirur-gisches Forum ‘84 für experimentelle undklinische Forschung. Berlin: Springer; 1984:69–72.

60. Just, M. Influence of various agents on exper-imental thrombosis in the rabbit. In: Wenzel,E., Hellstern, P., Morgenstern, E., Kohler,M., von Blohn, G., eds. Rationelle Ther-apie und Diagnose von hämorrhagischenund thrombophilen Diathesen. Stuttgart:Schattauer Verlag;1986:4.8–4.95.

61. Bevilacqua, C., Finesso, M., and Prosdocimi,M. (1991) Acute carotid artery occlusivethrombosis and its pharmacological preven-tion in the rabbit Thromb Res 62, 263–73.

62. Spokas, E.G., and Wun, T.C. (1992) Venousthrombosis produced in the vena cava of rab-bits by vascular damage and stasis J Pharma-col Toxicol Methods 27, 225–32.

63. Lyle, E.M., Fujita, T., Conner, M.W.,Connolly, T.M., Vlasuk, G.P., and Lynch,J.L., Jr. (1995) Effect of inhibitors of factorXa or platelet adhesion, heparin, and aspirinon platelet deposition in an atheroscleroticrabbit model of angioplasty injury J Pharma-col Toxicol Methods 33, 53–61.

64. Meng, K. (1975) Tierexperimentelle Unter-suchungen zur antithrombotischen Wirkungvon Acetylsalicylsäure Therap Ber 47,69–79.

65. Meng, K., and Seuter, F. (1977) Effectof acetylsalicylic acid on experimentallyinduced arterial thrombosis in rats Naunyn-Schmiedeberg’s Arch Pharmacol 301,115–9.

66. Seuter, F., Busse, W.D., Meng, K.,Hoffmeister, F., Moller, E., and Horstmann,H. (1979) The antithrombotic activity ofBAY g 6575 Arzneimittel-Forschung 29,54–9.

67. Salazar, A.E. (1961) Experimental myocar-dial infarction. Induction of coronary throm-bosis in the intact closed-chest dog Circ Res9, 1351–6.

68. Schumacher, W.A., Lee, E.C., and Lucchesi,B.R. (1985) Augmentation of streptokinase-induced thrombolysis by heparin and prosta-cyclin J Cardiovasc Pharmacol 7, 739–46.

69. van der Giessen, W.J., Harmsen, E.,de Tombe, P.P., Hugenholtz, P.G., andVerdouw, P.D. (1988) Coronary thromboly-sis with and without nifedipine in pigs BasicRes Cardiol 83, 258–67.

70. Shea, M.J., Driscoll, E.M., Romson, J.L.,Pitt, B., and Lucchesi, B.R. (1984) Effectof OKY-1581, a thromboxane synthetaseinhibitor, on coronary thrombosis in the con-scious dog Eur J Pharmacol 105, 285–91.

71. Simpson, P.J., Smith, C.B., Jr., Rosenthal, G.,and Lucchesi, B.R. (1986) Reduction in theincidence of thrombosis by the thromboxanesynthetase inhibitor CGS 13080 in a caninemodel of coronary artery injury J PharmacolExp Therap 238, 497–501.

72. Hook, B.G., Schumacher, W.A., Lee, D.L.,Jolly, S.R., and Lucchesi, B.R. (1985) Exper-imental coronary artery thrombosis in theabsence of thromboxane A2 synthesis: evi-dence for alternate pathways for coro-nary thrombosis J Cardiovasc Pharmacol 7,174–81.

73. Schumacher, W.A., and Lucchesi, B.R.(1983) Effect of the thromboxane syn-thetase inhibitor UK-37,248 (dazoxiben)upon platelet aggregation, coronary arterythrombosis and vascular reactivity J Pharma-col Exp Therap 227, 790–6.

74. Fitzgerald, D.J., Fragetta, J., and FitzGerald,G.A. (1988) Prostaglandin endoperoxidesmodulate the response to thromboxane syn-thase inhibition during coronary thrombosisJ Clin Invest 82, 1708–13.

75. Mickelson, J.K., Simpson, P.J., and Lucchesi,B.R. (1989) Antiplatelet monoclonal F(ab′)2antibody directed against the plateletGPIIb/IIIa receptor complex prevents coro-nary artery thrombosis in the canine heartJ Mol Cell Cardiol 21, 393–405.

76. Shebuski, R.J., Smith, J.M., Jr., Storer,B.L., Granett, J.R., and Bugelski, P.J.(1988) Influence of selective endoperox-ide/thromboxane A2 receptor antagonismwith sulotroban on lysis time and reocclu-sion rate after tissue plasminogen activator-induced coronary thrombolysis in the dogJ Pharmacol Exp Therap 246, 790–6.

77. van der Giessen, W.J., Zijlstra, F.J., Berk,L., and Verdouw, P.D. (1988) The effectof the thromboxane receptor antagonist BM13.177 on experimentally induced coronaryartery thrombosis in the pig Eur J Pharmacol147, 241–8.

78. Patterson, E., Eller, B.T., Abrams, G.D.,Vasiliades, J., and Lucchesi, B.R. (1983)Ventricular fibrillation in a consciouscanine preparation of sudden coronarydeath–prevention by short- and long-term


amiodarone administration Circulation 68,857–64.

79. Lefkovits, J., Malycky, J.L., Rao, J.S., Hart,C.E., Plow, E.F., Topol, E.J., and Nicolini,F.A. (1996) Selective inhibition of factorXa is more efficient than factor VIIa-tissuefactor complex blockade at facilitating coro-nary thrombolysis in the canine model J AmColl Cardiol 28, 1858–65.

80. Lynch, J.J., Jr., Sitko, G.R., Mellott, M.J.,Nutt, E.M., Lehman, E.D., Friedman, P.A.,Dunwiddie, C.T., and Vlasuk, G.P. (1994)Maintenance of canine coronary arterypatency following thrombolysis with frontloaded plus low dose maintenance conjunc-tive therapy. A comparison of factor Xa ver-sus thrombin inhibition Cardiovasc Res 28,78–85.

81. Nicolini, F.A., Lee, P., Rios, G., Kottke-Marchant, K., and Topol, E.J. (1994)Combination of platelet fibrinogen receptorantagonist and direct thrombin inhibitor atlow doses markedly improves thrombolysisCirculation 89, 1802–9.

82. Sitko, G.R., Ramjit, D.R., Stabilito, II,Lehman, D., Lynch, J.J., and Vlasuk, G.P.(1992) Conjunctive enhancement of enzy-matic thrombolysis and prevention of throm-botic reocclusion with the selective factor Xainhibitor, tick anticoagulant peptide. Com-parison to hirudin and heparin in a caninemodel of acute coronary artery thrombosisCirculation 85, 805–15.

83. Waxman, L., Smith, D.E., Arcuri, K.E., andVlasuk, G.P. (1990) Tick anticoagulant pep-tide (TAP) is a novel inhibitor of blood coag-ulation factor Xa Science 248, 593–6.

84. Lutz, B.R., Fulton, G.P., and Akers, R.P.(1951) White thromboembolism in the ham-ster cheek pouch after trauma, infection andneoplasia Circulation 3, 339–51.

85. Sawyer, P.N., and Pate, J.W. (1953) Bio-electric phenomena as an etiologic factor inintravascular thrombosis Am J Physiol 175,103–7.

86. Sawyer, P.N., Pate, J.W., and Weldon, C.S.(1953) Relations of abnormal and injuryelectric potential differences to intravascularthrombosis Am J Physiol 175, 108–12.

87. Hladovec, J. (1971) Experimental arterialthrombosis in rats with continuous registra-tion Thromb Diath Haemorrh 26, 407–10.

88. Philp, R.B., Francey, I., and Warren, B.A.(1978) Comparison of antithrombotic activ-ity of heparin, ASA, sulfinpyrazone and VK744 in a rat model of arterial thrombosisHaemostasis 7, 282–93.

89. Rote, W.E., Mu, D.X., Roncinske, R.A.,Frelinger, A.L., 3rd, and Lucchesi, B.R.

(1993) Prevention of experimental carotidartery thrombosis by applaggin J PharmacolExp Therap 267, 809–14.

90. Rote, W.E., Davis, J.H., Mousa, S.A.,Reilly, T.M., and Lucchesi, B.R. (1994)Antithrombotic effects of DMP 728, aplatelet GPIIb/IIIa receptor antagonist, in acanine model of arterial thrombosis J Cardio-vasc Pharmacol 23, 681–9.

91. Schumacher, W.A., Heran, C.L., andSteinbacher, T.E. (1996) Low-molecular-weight heparin (fragmin) and thrombinactive-site inhibitor (argatroban) comparedin experimental arterial and venous thrombo-sis and bleeding time J Cardiovasc Pharmacol28, 19–25.

92. Guarini, S. (1996) A highly reproduciblemodel of arterial thrombosis in rats J Phar-macol Toxicol Methods 35, 101–5.

93. Kurz, K.D., Main, B.W., and Sandusky, G.E.(1990) Rat model of arterial thrombosisinduced by ferric chloride Thromb Res 60,269–80.

94. Deschenes, I., Finkle, C.D., and Winocour,P.D. (1998) Effective use of BCH-2763,a new potent injectable direct thrombininhibitor, in combination with tissue plas-minogen activator (tPA) in a rat arterialthrombolysis model Thromb Haemost 80,186–91.

95. Elg, M., Gustafsson, D., and Carlsson, S.(1999) Antithrombotic effects and bleedingtime of thrombin inhibitors and warfarin inthe rat Thromb Res 94, 187–97.

96. Schumacher, W.A., Heran, C.L., Stein-bacher, T.E., Youssef, S., and Ogletree, M.L.(1993) Superior activity of a thrombox-ane receptor antagonist as compared withaspirin in rat models of arterial and venousthrombosis J Cardiovasc Pharmacol 22,526–33.

97. Roesken, F., Ruecker, M., Vollmar, B.,Boeckel, N., Morgenstern, E., and Menger,M.D. (1997) A new model for quantitativein vivo microscopic analysis of thrombus for-mation and vascular recanalisation: the ear ofthe hairless (hr/hr) mouse Thromb Haemost78, 1408–14.

98. Saniabadi, A.R., Umemura, K., Matsumoto,N., Sakuma, S., and Nakashima, M. (1995)Vessel wall injury and arterial thrombosisinduced by a photochemical reaction ThrombHaemost 73, 868–72.

99. Hokamura, K., Umemura, K., Nakamura,N., Watanabe, M., Takashima, T., andNakashima, M. (1998) Effect of lipo-pro-prostaglandin E1, AS-013 on rat inner earmicrocirculatory thrombosis ProstaglandinsLeukot Essent Fatty Acids 59, 203–7.

98 Mousa

100. Reimann-Hunziger, G. (1944) Über exper-imentelle Thrombose und ihre Behand-lung mit Heparin Schweiz Med Wschr 74,66–9.

101. Broersma, R.J., Kutcher, L.W., andHeminger, E.F. (1991) The effect ofthrombin inhibition in a rat arterialthrombosis model Thromb Res 64, 405–12.

102. van Giezen, J.J., Wahlund, G., Nerme, andAbrahamsson, T. (1997) The Fab-fragmentof a PAI-1 inhibiting antibody reducesthrombus size and restores blood flow ina rat model of arterial thrombosis ThrombHaemost 77, 964–9.

103. Shebuski, R.J., Storer, B.L., and Fujita, T.(1988) Effect of thromboxane synthase inhi-bition on the thrombolytic action of tissue-type plasminogen activator in a rabbit modelof peripheral arterial thrombosis Thromb Res52, 381–92.

104. Gold, H.K., Fallon, J.T., Yasuda, T.,Leinbach, R.C., Khaw, B.A., Newell, J.B.,Guerrero, J.L., Vislosky, F.M., Hoyng, C.F.,and Grossbard, E., et al. (1984) Coro-nary thrombolysis with recombinant humantissue-type plasminogen activator Circulation70, 700–7.

105. Collen, D., Stassen, J.M., and Verstraete, M.(1983) Thrombolysis with human extrinsic(tissue-type) plasminogen activator in rab-bits with experimental jugular vein throm-bosis. Effect of molecular form and doseof activator, age of the thrombus, androute of administration J Clin Invest 71,368–76.

106. Kopia, G.A., Kopaciewicz, L.J., and Ruffolo,R.R., Jr. (1988) Coronary thrombolysis withintravenous streptokinase in the anesthetizeddog: a dose-response study J Pharmacol ExpTherap 244, 956–62.

107. Gold, H.K., Leinbach, R.C., Garabedian,H.D., Yasuda, T., Johns, J.A., Grossbard,E.B., Palacios, I., and Collen, D. (1986)Acute coronary reocclusion after throm-bolysis with recombinant human tissue-type plasminogen activator: prevention bya maintenance infusion Circulation 73,347–52.

108. Gold, H.K., Coller, B.S., Yasuda, T., Saito,T., Fallon, J.T., Guerrero, J.L., Leinbach,R.C., Ziskind, A.A., and Collen, D. (1988)Rapid and sustained coronary artery recanal-ization with combined bolus injection ofrecombinant tissue-type plasminogen activa-tor and monoclonal antiplatelet GPIIb/IIIaantibody in a canine preparation Circulation77, 670–7.

109. Goldberg, R.K., Levine, S., and Fenster, P.E.(1985) Management of patients after throm-

bolytic therapy for acute myocardial infarc-tion Clin Cardiol 8, 455–9.

110. Coller, B.S., Peerschke, E.I., Scudder, L.E.,and Sullivan, C.A. (1983) A murine mon-oclonal antibody that completely blocks thebinding of fibrinogen to platelets produces athrombasthenic-like state in normal plateletsand binds to glycoproteins IIb and/or IIIaJ Clin Invest 72, 325–38.

111. Yasuda, T., Gold, H.K., Fallon, J.T.,Leinbach, R.C., Guerrero, J.L., Scudder,L.E., Kanke, M., Shealy, D., Ross, M.J., andCollen, D., et al. (1988) Monoclonal anti-body against the platelet glycoprotein (GP)IIb/IIIa receptor prevents coronary arteryreocclusion after reperfusion with recom-binant tissue-type plasminogen activator indogs J Clin Invest 81, 1284–91.

112. Weichert, W., and Breddin, H.K. (1988)Effect of low-molecular-weight heparins onlaser-induced thrombus formation in ratmesenteric vessels Haemostasis 18(Suppl 3),55–63.

113. Arfors, K.E., Dhall, D.P., Engeset, J., Hint,H., Matheson, N.A., and Tangen, O. (1968)Biolaser endothelial trauma as a means ofquantifying platelet activity in vivo Nature218, 887–8.

114. Herrmann, K.S. (1983) Platelet aggrega-tion induced in the hamster cheek pouch bya photochemical process with excited fluo-rescein isothiocyanate-dextran Microvasc Res26, 238–49.

115. Seiffge, D., and Kremer, E. (1984)Antithrombotic effects of pentoxifylline onlaser-induced thrombi in rat mesenteric arter-ies IRCS Med Sci 12, 91–2.

116. Seiffge, D., and Kremer, E. (1986) Influenceof ADP, blood flow velocity, and vessel diam-eter on the laser-induced thrombus forma-tion Thromb Res 42, 331–41.

117. Seiffge, D., and Weithmann, K.U. (1987)Surprising effects of the sequential adminis-tration of pentoxifylline and low dose acetyl-salicylic acid on thrombus formation ThrombRes 46, 371–83.

118. Weichert, W., Pauliks, V., and Breddin,H.K. (1983) Laser-induced thrombi in ratmesenteric vessels and antithrombotic drugsHaemostasis 13, 61–71.

119. Rosenblum, W.I., and El-Sabban, F. (1977)Platelet aggregation in the cerebral microcir-culation: effect of aspirin and other agentsCirc Res 40, 320–8.

120. Matsuno, H., Uematsu, T., Nagashima, S.,and Nakashima, M. (1991) Photochemicallyinduced thrombosis model in rat femoralartery and evaluation of effects of heparinand tissue-type plasminogen activator with


use of this model J Pharmacol Methods 25,303–17.

121. Stone, P., and Lord, J.W., Jr. (1951) Anexperimental study of the thrombogenicproperties of magnesium and magnesium-aluminum wire in the dog’s aorta Surgery 30,987–93.

122. Mellott, M.J., Stranieri, M.T., Sitko, G.R.,Stabilito, I.I., Lynch Jr, J.J., and Vlasuk,G.P. (1993) Enhancement of recombinanttissue plasminogen activator-induced reper-fusion by recombinant tick anticoagulantpeptide, a selective factor Xa inhibitor, in acanine model of femoral arterial thrombosisFibrinolysis 7, 195–202.

123. Rubsamen, K., and Hornberger, W. (1996)Prevention of early reocclusion after throm-bolysis of copper coil-induced thrombi in thecanine carotid artery: comparison of PEG-hirudin and unfractionated heparin ThrombHaemost 76, 105–10.

124. Kumada, T., Ishihara, M., Ogawa, H.,and Abiko, Y. (1980) Experimentalmodel of venous thrombosis in rats andeffect of some agents Thromb Res 18,189–203.

125. Bernat, A., Vallee, E., and Maffrand, J.(1986) A simple experimental model ofthrombolysis in the rat: effect of urokinaseand of the complex human plasminogen-streptokinase Thromb Res 44, 112.

126. Hergrueter, C.A., Handren, J., Kersh, R.,and May, J.W., Jr. (1988) Human recom-binant tissue type plasminogen activatorand its effect on microvascular thrombo-sis in the rabbit Plast Reconstruct Surg 81,418–24.

127. Jang, I.K., Gold, H.K., Ziskind, A.A., Fallon,J.T., Holt, R.E., Leinbach, R.C., May, J.W.,and Collen, D. (1989) Differential sensitivityof erythrocyte-rich and platelet-rich arterialthrombi to lysis with recombinant tissue-typeplasminogen activator. A possible explanationfor resistance to coronary thrombolysis Cir-culation 79, 920–8.

128. Jang, I.K., Gold, H.K., Ziskind, A.A.,Leinbach, R.C., Fallon, J.T., and Collen,D. (1990) Prevention of platelet-rich arterialthrombosis by selective thrombin inhibitionCirculation 81, 219–25.

129. Kaiser, B., and Fareed, J. (1996) Recom-binant full-length tissue factor pathwayinhibitor (TFPI) prevents thrombus forma-tion and rethrombosis after lysis in a rab-bit model of jugular vein thrombosis ThrombHaemost 76, 615–20.

130. Kaiser, B., Simon, A., and Markwardt, F.(1990) Antithrombotic effects of recombi-nant hirudin in experimental angioplasty and

intravascular thrombolysis Thromb Haemost63, 44–7.

131. Katsuragawa, M., Fujiwara, H., Kawamura,A., Htay, T., Yoshikuni, Y., Mori, K., andSasayama, S. (1993) An animal model ofcoronary thrombosis and thrombolysis—comparisons of vascular damage and throm-bus formation in the coronary and femoralarteries after balloon angioplasty Jpn Circ J57, 1000–6.

132. Kaiser, B., and Markwardt, F. (1986) Exper-imental studies on the antithrombotic actionof a highly effective synthetic thrombininhibitor Thromb Haemost 55, 194–6.

133. Heras, M., Chesebro, J.H., Penny, W.J.,Bailey, K.R., Lam, J.Y., Holmes, D.R.,Reeder, G.S., Badimon, L., and Fuster,V. (1988) Importance of adequate heparindosage in arterial angioplasty in a porcinemodel Circulation 78, 654–60.

134. Harker, L.A. (1987) Role of platelets andthrombosis in mechanisms of acute occlusionand restenosis after angioplasty Am J Cardiol60, 20B–8B.

135. Rowntree, L., and Shionoya, T. (1927) Stud-ies in extracorporeal thrombosis. I. A methodfor the direct observation of extracorporalthrombus formation. II. Thrombosis forma-tion in normal blood in the extracorporealvascular loop. III. Effects of certain antico-agulants (heparin and hirudin) on extracor-poreal thrombosis and on the mechanism ofthrombus formation J Exp Med 46, 7–26.

136. Knabb, R.M., Kettner, C.A., Timmermans,P.B., and Reilly, T.M. (1992) In vivo char-acterization of a new synthetic thrombininhibitor Thromb Haemost 67, 56–9.

137. Hara, T., Yokoyama, A., Tanabe, K., Ishi-hara, H., and Iwamoto, M. (1995) DX-9065a, an orally active, specific inhibitor offactor Xa, inhibits thrombosis without affect-ing bleeding time in rats Thromb Haemost 74,635–9.

138. Scott, N.A., Nunes, G.L., King, S.B., 3rd,Harker, L.A., and Hanson, S.R. (1994)Local delivery of an antithrombin inhibitsplatelet-dependent thrombosis Circulation90, 1951–5.

139. Best, C.H., Cowan, C., and Maclean, D.L.(1938) Heparin and the formation of whitethrombi J Physiol 92, 20–31.

140. Yokoyama, T., Kelly, A.B., Marzec, U.M.,Hanson, S.R., Kunitada, S., and Harker,L.A. (1995) Antithrombotic effects of orallyactive synthetic antagonist of activated fac-tor X in nonhuman primates Circulation 92,485–91.

141. Hollenbach, S., Wong, A., Ku, P., Needham,K., Lin, P.H., and Sinha, U. (1995) Efficacy

100 Mousa

of FXa inhibitors in a rabbit model of venousthrombosis Circulation 92, I486–7.

142. Hollenbach, S., Sinha, U., Lin, P.H.,Needham, K., Frey, L., Hancock, T., Wong,A., and Wolf, D. (1994) A comparative studyof prothrombinase and thrombin inhibitorsin a novel rabbit model of non-occlusivedeep vein thrombosis Thromb Haemost 71,357–62.

143. Gryglewski, R.J., Korbut, R., Ocetkiewicz,A., and Stachura, J. (1978) In vivo methodfor quantitation for anti-platelet potency ofdrugs Naunyn-Schmiedeberg’s Arch Pharma-col 302, 25–30.

144. Wessler, S. (1952) Studies in intravascularcoagulation. I. Coagulation changes in iso-lated venous segments J Clin Invest 31,1011–4.

145. Wessler, S. (1953) Studies in intravascularcoagulation. II. A comparison of the effectof dicumarol and heparin on clot formationin isolated venous segments J Clin Invest 32,650–4.

146. Wessler, S. (1955) Studies in intravascularcoagulation. III. The pathogenesis of serum-induced venous thrombosis J Clin Invest 34,647–51.

147. Wessler, S., Ballon, J.D., and Katz, J.H.(1957) Studies in intravascular coagulation.V. A distinction between the anticoagulantand antithrombotic effects of dicumarol NEngl J Med 256, 1223–5.

148. Wessler, S., and Morris, L.E. (1955) Stud-ies in intravascular coagulation. IV. The effectof heparin and dicumarol on serum-inducedvenous thrombosis Circulation 12, 553–6.

149. Fareed, J., Walenga, J.M., Kumar, A., andRock, A. (1985) A modified stasis thrombo-sis model to study the antithrombotic actionsof heparin and its fractions Semin ThrombHemost 11, 155–75.

150. Aronson, D.L., and Thomas, D.P. (1985)Experimental studies on venous thrombosis:effect of coagulants, procoagulants and vesselcontusion Thromb Haemost 54, 866–70.

151. Breddin, H.K. (1989) Thrombosis and Vir-chow’s triad: what is established? SeminThromb Hemost 15, 237–9.

152. Levi, M., Biemond, B.J., van Zonneveld,A.J., ten Cate, J.W., and Pannekoek, H.(1992) Inhibition of plasminogen activatorinhibitor-1 activity results in promotionof endogenous thrombolysis and inhibi-tion of thrombus extension in models ofexperimental thrombosis Circulation 85,305–12.

153. Biemond, B.J., Friederich, P.W., Levi, M.,Vlasuk, G.P., Buller, H.R., and ten Cate, J.W.(1996) Comparison of sustained antithrom-

botic effects of inhibitors of thrombin andfactor Xa in experimental thrombosis Circu-lation 93, 153–60.

154. Herbert, J.M., Bernat, A., Dol, F., Herault,J.P., Crepon, B., and Lormeau, J.C. (1996)DX 9065A a novel, synthetic, selective andorally active inhibitor of factor Xa: in vitroand in vivo studies J Pharmacol Exp Therap276, 1030–8.

155. Sato, K., Kawasaki, T., Hisamichi, N.,Taniuchi, Y., Hirayama, F., Koshio, H.,and Matsumoto, Y. (1998) Antithromboticeffects of YM-60828, a newly synthesizedfactor Xa inhibitor, in rat thrombosis modelsand its effects on bleeding time Br J Pharma-col 123, 92–6.

156. Yamazaki, M., Asakura, H., Aoshima,K., Saito, M., Jokaji, H., Uotani, C.,Kumabashiri, I., Morishita, E., Ikeda, T., andMatsuda, T. (1994) Effects of DX-9065a, anorally active, newly synthesized and specificinhibitor of factor Xa, against experimentaldisseminated intravascular coagulation in ratsThromb Haemost 72, 392–6.

157. Fu, K., Izquierdo, R., Vandevender, D.,Warpeha, R.L., Wolf, H., and Fareed, J.(1997) Topical application of low molecularweight heparin in a rabbit traumatic anasto-mosis model Thromb Res 86, 355–61.

158. Korompilias, A.V., Chen, L.E., Seaber, A.V.,and Urbaniak, J.R. (1997) Antithromboticpotencies of enoxaparin in microvascularsurgery: influence of dose and administrationmethods on patency rate of crushed arterialanastomoses J Hand Surg 22, 540–6.

159. Stockmans, F., Stassen, J.M., Vermylen, J.,Hoylaerts, M.F., and Nystrom, A. (1997) Atechnique to investigate mural thrombus for-mation in small arteries and veins: I. Compar-ative morphometric and histological analysisAnn Plast Surg 38, 56–62.

160. van Wyk, V., Neethling, W.M., Badenhorst,P.N., and Kotze, H.F. (1998) r-Hirudininhibits platelet-dependent thrombosis dur-ing cardiopulmonary bypass in baboonsJ Cardiovasc Surg 39, 633–9.

161. Dewanjee, M.K., Wu, S.M., Kapadvanjwala,M., De, D., Dewanjee, S., Novak, S.,Hsu, L.C., Perryman, R.A., Serafini, A.N.,Sfakianakis, G.N., Duncan, R.C., Dietrich,W.D., and Horton, A.F. (1996) Reduc-tion of platelet thrombi and emboli byL-arginine during cardiopulmonary bypassin a pig model J Thromb Thrombolysis 3,343–60.

162. Henny, C.P., Ten Cate, H., Ten Cate, J.W.,Moulijn, A.C., Sie, T.H., Warren, P., andBuller, H.R. (1985) A randomized blindstudy comparing standard heparin and a


new low molecular weight heparinoid in car-diopulmonary bypass surgery in dogs J LabClin Med 106, 187–96.

163. Murray, W. A preliminary study of lowmolecular weight heparin in aortocoronaybypass surgery. In: Murray, W., ed. LowMolecular Weight Heparin in Surgical Prac-tice (Master of Surgery Thesis). London:University of London;1985:266.

164. Badimon, L., and Badimon, J.J. (1989)Mechanisms of arterial thrombosis in non-parallel streamlines: platelet thrombi grow onthe apex of stenotic severely injured vesselwall. Experimental study in the pig modelJ Clin Invest 84, 1134–44.

165. Wysokinski, W., McBane, R., Chesebro, J.H.,and Owen, W.G. (1996) Reversibility ofplatelet thrombosis in vivo. Quantitativeanalysis in porcine carotid arteries ThrombHaemost 76, 1108–13.

166. Dangas, G., Badimon, J.J., Coller, B.S.,Fallon, J.T., Sharma, S.K., Hayes, R.M.,Meraj, P., Ambrose, J.A., and Marmur, J.D.(1998) Administration of abciximab duringpercutaneous coronary intervention reducesboth ex vivo platelet thrombus formation andfibrin deposition: implications for a potentialanticoagulant effect of abciximab ArteriosclerThromb Vasc Biol 18, 1342–9.

167. Ørvim, U., Barstad R.M., Vlasuk., G.P., andSakariassen, K.S. (1995) Effect of selectivefactor Xa inhibition on arterial thrombusformation triggered by tissue factor/factorVIIa or collagen in an ex vivo modelof shear-dependent human thrombogenesisArterioscler Thromb Vasc Biol 15, 2188–94.

168. Angelillo-Scherrer, A., de Frutos, P.,Aparicio, C., Melis, E., Savi, P., Lupu, F.,Arnout, J., Dewerchin, M., Hoylaerts, M.,Herbert, J., Collen, D., Dahlback, B., andCarmeliet, P. (2001) Deficiency or inhibitionof Gas6 causes platelet dysfunction and pro-tects mice against thrombosis Nat Med 7,215–21.

169. Yang, J., Wu, J., Kowalska, M.A., Dalvi,A., Prevost, N., O’Brien, P.J., Manning, D.,Poncz, M., Lucki, I., Blendy, J.A., and Brass,L.F. (2000) Loss of signaling through the Gprotein, Gz, results in abnormal platelet acti-vation and altered responses to psychoactivedrugs Proc Natl Acad Sci USA 97, 9984–9.

170. Volkl, K.P., and Dierichs, R. (1986) Effectof intravenously injected collagenase on theconcentration of circulating platelets in ratsThromb Res 42, 11–20.

171. Oyekan, A.O., and Botting, J.H. (1986) Aminimally invasive technique for the studyof intravascular platelet aggregation in anes-thetized rats J Pharmacol Methods 15, 271–7.

172. Smith, D., Sanjar, S., Herd, C., and Morley,J. (1989) In vivo method for the assessmentof platelet accumulation J Pharmacol Methods21, 45–59.

173. Dottl, K., and Ripke, O. Blutgerinnungund Blutungszeit. In: Medizin und Chemie.Leverkusen: Bayer;1936:267–73.

174. Dejana, E., Callioni, A., Quintana, A., andde Gaetano, G. (1979) Bleeding time inlaboratory animals. II—A comparison of dif-ferent assay conditions in rats Thromb Res 15,191–7.

175. Minsker, D.H., and Kling, P.J. (1977) Bleed-ing time in rats is prolonged by aspirinThromb Res 10, 619–22.

176. Stella, L., Donati, M.B., and de Gaetano, G.(1975) Bleeding time in laboratory animalsI. Aspirin does not prolong bleeding time inrats Thromb Res 7, 709–16.

177. Butler, K.D., Maguire, E.D., Smith, J.R.,Turnbull, A.A., Wallis, R.B., and White, A.M.(1982) Prolongation of rat tail bleeding timecaused by oral doses of a thromboxane syn-thetase inhibitor which have little effect onplatelet aggregation Thromb Haemost 47,46–9.

178. Zawilska, K.M., Born, G.V., and Begent,N.A. (1982) Effect of ADP-utilizingenzymes on the arterial bleeding time in ratsand rabbits Br J Haematol 50, 317–25.

179. Forsythe, L.T., and Willis, S.E. (1989) Eval-uating oral mucosa bleeding times in healthydogs using a spring-loaded device Can Veter J30, 344–5.

180. Klement, P., Liao, P., Hirsh, J., Johnston, M.,and Weitz, J.I. (1998) Hirudin causes morebleeding than heparin in a rabbit ear bleedingmodel J Lab Clin Med 132, 181–5.

181. Brassard, J.A., and Meyers, K.M. (1991)Evaluation of the buccal bleeding time andplatelet glass bead retention as assays ofhemostasis in the dog: the effects of acetylsal-icylic acid, warfarin and von Willebrand fac-tor deficiency Thromb Haemost 65, 191–5.

182. Lind, S.E. (1991) The bleeding time doesnot predict surgical bleeding Blood 77,2547–52.

183. Giles, A.R., Tinlin, S., and Greenwood, R.(1982) A canine model of hemophilic (fac-tor VIII: C deficiency) bleeding Blood 60,727–30.

184. Carmeliet, P., and Collen, D. Newdevelopments in the molecular biology ofcoagulation and fibrinolysis. In: Hand-book of Experimental Pharmacology, Vol132, Antithrombotics. Berlin: Springer-Verlag;1999:41–76.

185. Pearson, J., and Ginsburg, D. Use of trans-genic mice in the study of thrombosis and

102 Mousa

hemostasis. In: Handbook of Experimen-tal Pharmacology, Vol 132, Antithrombotics.Berlin: Springer-Verlag; 1999:157–74.

186. Bi, L., Sarkar, R., Naas, T., Lawler, A.M.,Pain, J., Shumaker, S.L., Bedian, V., andKazazian, H.H., Jr. (1996) Further charac-terization of factor VIII-deficient mice cre-ated by gene targeting: RNA and proteinstudies Blood 88, 3446–50.

187. Denis, C., Methia, N., Frenette, P.S.,Rayburn, H., Ullman-Cullere, M., Hynes,R.O., and Wagner, D.D. (1998) A mousemodel of severe von Willebrand disease:defects in hemostasis and thrombosis ProcNatl Acad Sci USA 95, 9524–9.

188. Hodivala-Dilke, K.M., McHugh, K.P.,Tsakiris, D.A., Rayburn, H., Crowley, D.,Ullman-Cullere, M., Ross, F.P., Coller, B.S.,Teitelbaum, S., and Hynes, R.O. (1999)Beta3-integrin-deficient mice are a model forGlanzmann thrombasthenia showing placen-tal defects and reduced survival J Clin Invest103, 229–38.

189. Wang, L., Zoppe, M., Hackeng, T.M.,Griffin, J.H., Lee, K.F., and Verma, I.M.(1997) A factor IX-deficient mouse modelfor hemophilia B gene therapy Proc NatlAcad Sci USA 94, 11563–6.

190. Offermanns, S., Toombs, C.F., Hu, Y.H.,and Simon, M.I. (1997) Defective plateletactivation in G alpha(q)-deficient miceNature 389, 183–6.

191. Law, D.A., DeGuzman, F.R., Heiser, P.,Ministri-Madrid, K., Killeen, N., and Phillips,D.R. (1999) Integrin cytoplasmic tyrosinemotif is required for outside-in alphaIIbbeta3signalling and platelet function Nature 401,808–11.

192. Kahn, M.L., Zheng, Y.W., Huang, W.,Bigornia, V., Zeng, D., Moff, S., Farese,R.V., Jr., Tam, C., and Coughlin, S.R. (1998)A dual thrombin receptor system for plateletactivation Nature 394, 690–4.

193. Leon, C., Hechler, B., Freund, M., Eckly, A.,Vial, C., Ohlmann, P., Dierich, A., LeMeur,M., Cazenave, J.P., and Gachet, C. (1999)Defective platelet aggregation and increasedresistance to thrombosis in purinergicP2Y(1) receptor-null mice J Clin Invest 104,1731–7.

194. Subramaniam, M., Frenette, P.S., Saffaripour,S., Johnson, R.C., Hynes, R.O., and Wagner,D.D. (1996) Defects in hemostasis inP-selectin-deficient mice Blood 87, 1238–42.

195. Thomas, D.W., Mannon, R.B., Mannon,P.J., Latour, A., Oliver, J.A., Hoffman, M.,Smithies, O., Koller, B.H., and Coffman,T.M. (1998) Coagulation defects and alteredhemodynamic responses in mice lacking

receptors for thromboxane A2 J Clin Invest102, 1994–2001.

196. Bugge, T.H., Flick, M.J., Daugherty, C.C.,and Degen, J.L. (1995) Plasminogen defi-ciency causes severe thrombosis but is com-patible with development and reproductionGenes Develop 9, 794–807.

197. Ploplis, V.A., Carmeliet, P., Vazirzadeh, S.,Van Vlaenderen, I., Moons, L., Plow, E.F.,and Collen, D. (1995) Effects of disrup-tion of the plasminogen gene on thrombosis,growth, and health in mice Circulation 92,2585–93.

198. Carmeliet, P., Schoonjans, L., Kieckens, L.,Ream, B., Degen, J., Bronson, R., De Vos,R., van den Oord, J.J., Collen, D., andMulligan, R.C. (1994) Physiological conse-quences of loss of plasminogen activator genefunction in mice Nature 368, 419–24.

199. Christie, P.D., Edelberg, J.M., Picard, M.H.,Foulkes, A.S., Mamuya, W., Weiler-Guettler,H., Rubin, R.H., Gilbert, P., and Rosenberg,R.D. (1999) A murine model of myocardialmicrovascular thrombosis J Clin Invest 104,533–9.

200. Carmeliet, P., Stassen, J.M., Schoonjans, L.,Ream, B., van den Oord, J.J., De Mol, M.,Mulligan, R.C., and Collen, D. (1993) Plas-minogen activator inhibitor-1 gene-deficientmice. II. Effects on hemostasis, thrombosis,and thrombolysis J Clin Invest 92, 2756–60.

201. Bugge, T.H., Xiao, Q., Kombrinck, K.W.,Flick, M.J., Holmback, K., Danton, M.J.,Colbert, M.C., Witte, D.P., Fujikawa, K.,Davie, E.W., and Degen, J.L. (1996) Fatalembryonic bleeding events in mice lackingtissue factor, the cell-associated initiator ofblood coagulation Proc Natl Acad Sci USA93, 6258–63.

202. Carmeliet, P., Mackman, N., Moons, L.,Luther, T., Gressens, P., Van Vlaenderen,I., Demunck, H., Kasper, M., Breier, G.,Evrard, P., Muller, M., Risau, W., Edgington,T., and Collen, D. (1996) Role of tissue fac-tor in embryonic blood vessel developmentNature 383, 73–5.

203. Toomey, J.R., Kratzer, K.E., Lasky, N.M.,Stanton, J.J., and Broze, G.J., Jr. (1996) Tar-geted disruption of the murine tissue factorgene results in embryonic lethality Blood 88,1583–7.

204. Huang, Z.F., Higuchi, D., Lasky, N., andBroze, G.J., Jr. (1997) Tissue factor pathwayinhibitor gene disruption produces intrauter-ine lethality in mice Blood 90, 944–51.

205. Healy, A.M., Rayburn, H.B., Rosenberg,R.D., and Weiler, H. (1995) Absence ofthe blood-clotting regulator thrombomod-ulin causes embryonic lethality in mice before


development of a functional cardiovascularsystem Proc Natl Acad Sci USA 92, 850–4.

206. Connolly, A.J., Ishihara, H., Kahn, M.L.,Farese, R.V., Jr., and Coughlin, S.R. (1996)Role of the thrombin receptor in develop-ment and evidence for a second receptorNature 381, 516–9.

207. Cui, J., O’Shea, K.S., Purkayastha, A.,Saunders, T.L., and Ginsburg, D. (1996)Fatal haemorrhage and incomplete block toembryogenesis in mice lacking coagulationfactor V Nature 384, 66–8.

208. Lin, H.F., Maeda, N., Smithies, O., Straight,D.L., and Stafford, D.W. (1997) A coagu-lation factor IX-deficient mouse model forhuman hemophilia B Blood 90, 3962–6.

209. Kung, S.H., Hagstrom, J.N., Cass, D., Tai,S.J., Lin, H.F., Stafford, D.W., and High,K.A. (1998) Human factor IX corrects thebleeding diathesis of mice with hemophilia BBlood 91, 784–90.

210. Evans, J.P., Brinkhous, K.M., Brayer, G.D.,Reisner, H.M., and High, K.A. (1989)Canine hemophilia B resulting from a pointmutation with unusual consequences ProcNatl Acad Sci USA 86, 10095–9.

211. Herzog, R.W., Yang, E.Y., Couto, L.B.,Hagstrom, J.N., Elwell, D., Fields, P.A.,Burton, M., Bellinger, D.A., Read, M.S.,Brinkhous, K.M., Podsakoff, G.M., Nichols,T.C., Kurtzman, G.J., and High, K.A.(1999) Long-term correction of caninehemophilia B by gene transfer of bloodcoagulation factor IX mediated by adeno-associated viral vector Nat Med 5, 56–63.

212. Kawasaki T., Dewerchin M., Lijnen H.R.,Vermylen J., Hoylaerts M.F. (2000) Vascu-lar release of plasminogen activator inhibitor-1 impairs fibrinolysis during acute arterialthrombosis in mice. Blood 96, 153–60.

213. Waugh, J.M., Yuksel, E., Li, J., Kuo, M.D.,Kattash, M., Saxena, R., Geske, R., Thung,S.N., Shenaq, S.M., and Woo, S.L. (1999)Local overexpression of thrombomodulin forin vivo prevention of arterial thrombosis in arabbit model Circ Res 84, 84–92.

214. Waugh, J.M., Kattash, M., Li, J., Yuksel,E., Kuo, M.D., Lussier, M., Weinfeld, A.B.,Saxena, R., Rabinovsky, E.D., Thung, S.,Woo, S.L., and Shenaq, S.M. (1999) Genetherapy to promote thromboresistance: localoverexpression of tissue plasminogen activa-tor to prevent arterial thrombosis in an invivo rabbit model Proc Natl Acad Sci USA96, 1065–70.

215. Zoldhelyi, P., McNatt, J., Shelat, H.S.,Yamamoto, Y., Chen, Z.Q., and Willerson,J.T. (2000) Thromboresistance of balloon-injured porcine carotid arteries after local

gene transfer of human tissue factor pathwayinhibitor Circulation 101, 289–95.

216. Vassalli, G., and Dichek, D.A. (1997) Genetherapy for arterial thrombosis CardiovascRes 35, 459–69.

217. Suh, T.T., Holmback, K., Jensen, N.J.,Daugherty, C.C., Small, K., Simon, D.I.,Potter, S., and Degen, J.L. (1995) Resolu-tion of spontaneous bleeding events but fail-ure of pregnancy in fibrinogen-deficient miceGenes Develop 9, 2020–33.

218. Yang, T.L., Cui, J., Taylor, J.M., Yang, A.,Gruber, S.B., and Ginsburg, D. (2000) Res-cue of fatal neonatal hemorrhage in factor Vdeficient mice by low level transgene expres-sion Thromb Haemost 83, 70–7.

219. Rosen, E.D., Chan, J.C., Idusogie, E.,Clotman, F., Vlasuk, G., Luther, T., Jalbert,L.R., Albrecht, S., Zhong, L., Lissens, A.,Schoonjans, L., Moons, L., Collen, D.,Castellino, F.J., and Carmeliet, P. (1997)Mice lacking factor VII develop normally butsuffer fatal perinatal bleeding Nature 390,290–4.

220. Bi, L., Lawler, A.M., Antonarakis, S.E.,High, K.A., Gearhart, J.D., and Kazazian,H.H., Jr. (1995) Targeted disruption of themouse factor VIII gene produces a model ofhaemophilia A Nat Genet 10, 119–21.

221. Kundu, R.K., Sangiorgi, F., Wu, L.Y.,Kurachi, K., Anderson, W.F., Maxson, R.,and Gordon, E.M. (1998) Targeted inactiva-tion of the coagulation factor IX gene causeshemophilia B in mice Blood 92, 168–74.

222. Dewerchin, M., Liang, Z., Moons, L.,Carmeliet, P., Castellino, F.J., Collen, D.,and Rosen, E.D. (2000) Blood coagulationfactor X deficiency causes partial embryoniclethality and fatal neonatal bleeding in miceThromb Haemost 83, 185–90.

223. Gailani, D., Lasky, N.M., and Broze, G.J.,Jr. (1997) A murine model of factor XI defi-ciency Blood Coagul Fibrinolysis 8, 134–44.

224. Toomey, J.R., Kratzer, K.E., Lasky, N.M.,and Broze, G.J., Jr. (1997) Effect of tissuefactor deficiency on mouse and tumor devel-opment Proc Natl Acad Sci USA 94, 6922–6.

225. Darrow, A.L., Fung-Leung, W.P., Ye, R.D.,Santulli, R.J., Cheung, W.M., Derian, C.K.,Burns, C.L., Damiano, B.P., Zhou, L.,Keenan, C.M., Peterson, P.A., and Andrade-Gordon, P. (1996) Biological consequencesof thrombin receptor deficiency in miceThromb Haemost 76, 860–6.

226. Healy, A.M., Hancock, W.W., Christie,P.D., Rayburn, H.B., and Rosenberg, R.D.(1998) Intravascular coagulation activationin a murine model of thrombomodulindeficiency: effects of lesion size, age, and

104 Mousa

hypoxia on fibrin deposition Blood 92,4188–97.

227. Weiler-Guettler, H., Christie, P.D.,Beeler, D.L., Healy, A.M., Hancock,W.W., Rayburn, H., Edelberg, J.M., andRosenberg, R.D. (1998) A targeted pointmutation in thrombomodulin generatesviable mice with a prethrombotic state J ClinInvest 101, 1983–91.

228. Jalbert, L.R., Rosen, E.D., Moons, L.,Chan, J.C., Carmeliet, P., Collen, D., andCastellino, F.J. (1998) Inactivation of thegene for anticoagulant protein C causes lethalperinatal consumptive coagulopathy in miceJ Clin Invest 102, 1481–8.

229. Lijnen, H.R., Okada, K., Matsuo, O., Collen,D., and Dewerchin, M. (1999) Alpha2-antiplasmin gene deficiency in mice is asso-ciated with enhanced fibrinolytic potentialwithout overt bleeding Blood 93, 2274–81.

230. Eitzman, D.T., McCoy, R.D., Zheng, X.,Fay, W.P., Shen, T., Ginsburg, D., andSimon, R.H. (1996) Bleomycin-induced pul-monary fibrosis in transgenic mice that eitherlack or overexpress the murine plasminogenactivator inhibitor-1 gene J Clin Invest 97,232–7.

231. Erickson, L.A., Fici, G.J., Lund, J.E., Boyle,T.P., Polites, H.G., and Marotti, K.R. (1990)Development of venous occlusions in micetransgenic for the plasminogen activatorinhibitor-1 gene Nature 346, 74–6.

232. Pinsky, D.J., Liao, H., Lawson, C.A., Yan,S.F., Chen, J., Carmeliet, P., Loskutoff, D.J.,and Stern, D.M. (1998) Coordinated induc-tion of plasminogen activator inhibitor-1(PAI-1) and inhibition of plasminogen acti-vator gene expression by hypoxia promotespulmonary vascular fibrin deposition J ClinInvest 102, 919–28.

233. Eitzman, D.T., Westrick, R.J., Nabel, E.G.,and Ginsburg, D. (2000) Plasminogen acti-vator inhibitor-1 and vitronectin promotevascular thrombosis in mice Blood 95,577–80.

234. Zheng, X., Saunders, T.L., Camper, S.A.,Samuelson, L.C., and Ginsburg, D. (1995)Vitronectin is not essential for normal mam-malian development and fertility Proc NatlAcad Sci USA 92, 12426–30.

235. Heckel, J.L., Sandgren, E.P., Degen, J.L.,Palmiter, R.D., and Brinster, R.L. (1990)Neonatal bleeding in transgenic miceexpressing urokinase-type plasminogen acti-vator Cell 62, 447–56.

236. Bugge, T.H., Flick, M.J., Danton, M.J.,Daugherty, C.C., Romer, J., Dano, K.,Carmeliet, P., Collen, D., and Degen, J.L.(1996) Urokinase-type plasminogen activa-

tor is effective in fibrin clearance in theabsence of its receptor or tissue-type plas-minogen activator Proc Natl Acad Sci USA93, 5899–904.

237. Bugge, T.H., Suh, T.T., Flick, M.J.,Daugherty, C.C., Romer, J., Solberg, H.,Ellis, V., Dano, K., and Degen, J.L. (1995)The receptor for urokinase-type plasminogenactivator is not essential for mouse develop-ment or fertility J Biol Chem 270, 16886–94.

238. Dewerchin, M., Nuffelen, A.V., Wallays,G., Bouche, A., Moons, L., Carmeliet, P.,Mulligan, R.C., and Collen, D. (1996) Gen-eration and characterization of urokinasereceptor-deficient mice J Clin Invest 97,870–8.

239. Piguet, P.F., Da Laperrousaz, C., Vesin, C.,Tacchini-Cottier, F., Senaldi, G., and Grau,G.E. (2000) Delayed mortality and atten-uated thrombocytopenia associated withsevere malaria in urokinase- and urokinasereceptor-deficient mice Infect Immun 68,3822–9.

240. Ware, J., Russell, S., and Ruggeri, Z.M.(2000) Generation and rescue of a murinemodel of platelet dysfunction: the Bernard-Soulier syndrome Proc Natl Acad Sci USA97, 2803–8.

241. Ramakrishnan, V., Reeves, P.S., DeGuzman,F., Deshpande, U., Ministri-Madrid, K.,DuBridge, R.B., and Phillips, D.R. (1999)Increased thrombin responsiveness inplatelets from mice lacking glycoprotein VProc Natl Acad Sci USA 96, 13336–41.

242. Tronik-Le Roux, D., Roullot, V., Poujol, C.,Kortulewski, T., Nurden, P., and Marguerie,G. (2000) Thrombasthenic mice generatedby replacement of the integrin alpha(IIb)gene: demonstration that transcriptional acti-vation of this megakaryocytic locus precedeslineage commitment Blood 96, 1399–408.

243. McHugh, K.P., Hodivala-Dilke, K., Zheng,M.H., Namba, N., Lam, J., Novack, D.,Feng, X., Ross, F.P., Hynes, R.O., andTeitelbaum, S.L. (2000) Mice lacking beta3integrins are osteosclerotic because of dys-functional osteoclasts J Clin Invest 105,433–40.

244. Nieswandt, B., Brakebusch, C., Bergmeier,W., Schulte, V., Bouvard, D., Mokhtari-Nejad, R., Lindhout, T., Heemskerk, J.W.,Zirngibl, H., and Fassler, R. (2001) Glyco-protein VI but not alpha2beta1 integrin isessential for platelet interaction with collagenEMBO J 20, 2120–30.

245. Ni, H., Denis, C.V., Subbarao, S., Degen,J.L., Sato, T.N., Hynes, R.O., and Wagner,D.D. (2000) Persistence of platelet thrombusformation in arterioles of mice lacking both


von Willebrand factor and fibrinogen J ClinInvest 106, 385–92.

246. Murata, T., Ushikubi, F., Matsuoka, T.,Hirata, M., Yamasaki, A., Sugimoto, Y.,Ichikawa, A., Aze, Y., Tanaka, T., Yoshida,N., Ueno, A., Oh-ishi, S., and Narumiya, S.(1997) Altered pain perception and inflam-matory response in mice lacking prostacyclinreceptor Nature 388, 678–82.

247. Duncan, G.S., Andrew, D.P., Takimoto, H.,Kaufman, S.A., Yoshida, H., Spellberg, J.,Luis de la Pompa, J., Elia, A., Wakeham,A., Karan-Tamir, B., Muller, W.A., Senaldi,G., Zukowski, M.M., and Mak, T.W.(1999) Genetic evidence for functionalredundancy of platelet/endothelial celladhesion molecule-1 (PECAM-1): CD31-deficient mice reveal PECAM-1-dependentand PECAM-1-independent functionsJ Immunol 162, 3022–30.

248. Mahooti, S., Graesser, D., Patil, S., Newman,P., Duncan, G., Mak, T., and Madri, J.A.(2000) PECAM-1 (CD31) expression mod-ulates bleeding time in vivo Am J Pathol 157,75–81.

249. Huang, L., Kuo, Y.M., and Gitschier, J.(1999) The pallid gene encodes a novel,syntaxin 13-interacting protein involved inplatelet storage pool deficiency Nat Genet23, 329–32.

250. Ohlmann, P., Eckly, A., Freund, M.,Cazenave, J.P., Offermanns, S., and Gachet,C. (2000) ADP induces partial platelet aggre-gation without shape change and potentiatescollagen-induced aggregation in the absenceof Galphaq Blood 96, 2134–9.

251. Wang, D., Feng, J., Wen, R., Marine, J.C.,Sangster, M.Y., Parganas, E., Hoffmeyer,A., Jackson, C.W., Cleveland, J.L., Murray,P.J., and Ihle, J.N. (2000) PhospholipaseCgamma2 is essential in the functions of Bcell and several Fc receptors Immunity 13,25–35.

252. Enjyoji, K., Sevigny, J., Lin, Y., Frenette,P.S., Christie, P.D., Esch, J.S., 2nd, Imai,M., Edelberg, J.M., Rayburn, H., Lech, M.,Beeler, D.L., Csizmadia, E., Wagner, D.D.,Robson, S.C., and Rosenberg, R.D. (1999)Targeted disruption of cd39/ATP diphos-phohydrolase results in disordered hemosta-sis and thromboregulation Nat Med 5,1010–7.

253. Massberg, S., Sausbier, M., Klatt, P., Bauer,M., Pfeifer, A., Siess, W., Fassler, R.,Ruth, P., Krombach, F., and Hofmann, F.(1999) Increased adhesion and aggregationof platelets lacking cyclic guanosine 3′,5′-monophosphate kinase I J Exp Med 189,1255–64.

254. Aszodi, A., Pfeifer, A., Ahmad, M., Glauner,M., Zhou, X.H., Ny, L., Andersson,K.E., Kehrel, B., Offermanns, S., andFassler, R. (1999) The vasodilator-stimulatedphosphoprotein (VASP) is involved incGMP- and cAMP-mediated inhibition ofagonist-induced platelet aggregation, butis dispensable for smooth muscle functionEMBO J 18, 37–48.

255. Hauser, W., Knobeloch, K.P., Eigenthaler,M., Gambaryan, S., Krenn, V., Geiger, J.,Glazova, M., Rohde, E., Horak, I., Walter,U., and Zimmer, M. (1999) Megakaryocytehyperplasia and enhanced agonist-inducedplatelet activation in vasodilator-stimulatedphosphoprotein knockout mice Proc NatlAcad Sci USA 96, 8120–5.

256. Chen, X.S., Sheller, J.R., Johnson, E.N.,and Funk, C.D. (1994) Role of leukotrienesrevealed by targeted disruption of the5-lipoxygenase gene Nature 372, 179–82.

257. Johnson, E.N., Brass, L.F., and Funk,C.D. (1998) Increased platelet sensitivityto ADP in mice lacking platelet-type 12-lipoxygenase Proc Nat Acad Sci USA 95,3100–5.

258. Lawler, J., Sunday, M., Thibert, V.,Duquette, M., George, E.L., Rayburn, H.,and Hynes, R.O. (1998) Thrombospondin-1 is required for normal murine pulmonaryhomeostasis and its absence causes pneumo-nia J Clin Invest 101, 982–92.

259. Schumacher, W.A., Steinbacher, T.E., Megill,J.R., and Durham, S.K. (1996) A fer-ret model of electrical-induction of arterialthrombosis that is sensitive to aspirin J Phar-macol Toxicol Methods 35, 3–10.

260. Sloan, M.A., and Gore, J.M. (1992) Ischemicstroke and intracranial hemorrhage follow-ing thrombolytic therapy for acute myocar-dial infarction: a risk-benefit analysis Am JCardiol 69, 21A–38A.

261. Stieg, P.E., and Kase, C.S. (1998) Intracra-nial hemorrhage: diagnosis and emergencymanagement Neurolog Clin 16, 373–90.

262. Bernardi, M.M., Califf, R.M., Kleiman, N.,Ellis, S.G., and Topol, E.J. (1993) Lackof usefulness of prolonged bleeding timesin predicting hemorrhagic events in patientsreceiving the 7E3 glycoprotein IIb/IIIaplatelet antibody. The TAMI Study GroupAm J Cardiol 72, 1121–5.

263. Bick, R.L. (1995) Laboratory evaluation ofplatelet dysfunction Clin Lab Med 15, 1–38.

264. Rodgers, R.P., and Levin, J. (1990) A crit-ical reappraisal of the bleeding time SeminThromb Hemost 16, 1–20.

265. (1996) A comparison of recombinant hirudinwith heparin for the treatment of acute

106 Mousa

coronary syndromes. The global use ofstrategies to open occluded coronary arter-ies (GUSTO) IIb investigators N Engl J Med335, 775–82.

266. Antman, E.M. (1996) Hirudin in acutemyocardial infarction. Thrombolysisand thrombin inhibition in myocardialinfarction (TIMI) 9B trial Circulation 94,911–21.

267. Cox, D., Motoyama, Y., Seki, J., Aoki, T.,Dohi, M., and Yoshida, K. (1992) Pentami-dine: a non-peptide GPIIb/IIIa antagonist–in vitro studies on platelets from humansand other species Thromb Haemost 68,731–6.

268. Bostwick, J.S., Kasiewski, C.J., Chu, V.,Klein, S.I., Sabatino, R.D., Perrone, M.H.,Dunwiddie, C.T., Cook, J.J., and Leadley,R.J., Jr. (1996) Anti-thrombotic activityof RG13965, a novel platelet fibrino-gen receptor antagonist Thromb Res 82,495–507.

269. Cook, N.S., Zerwes, H.G., Tapparelli, C.,Powling, M., Singh, J., Metternich, R., andHagenbach, A. (1993) Platelet aggregationand fibrinogen binding in human, rhesusmonkey, guinea-pig, hamster and rat blood:activation by ADP and a thrombin recep-tor peptide and inhibition by glycoproteinIIb/IIIa antagonists Thromb Haemost 70,531–9.

270. Panzer-Knodle, S., Taite, B.B., Mehrotra,D.V., Nicholson, N.S., and Feigen, L.P.(1993) Species variation in the effect of gly-coprotein IIb/IIIa antagonists on inhibitionof platelet aggregation J Pharmacol ToxicolMethods 30, 47–53.

271. Cook, J.J., Holahan, M.A., Lyle, E.A.,Ramjit, D.R., Sitko, G.R., Stranieri, M.T.,Stupienski, R.F., 3rd, Wallace, A.A., Hand,E.L., Gehret, J.R., Kothstein, T., Drag,M.D., McCormick, G.Y., Perkins, J.J., Ihle,N.C., Duggan, M.E., Hartman, G.D.,Gould, R.J., and Lynch, J.J., Jr. (1996) Non-peptide glycoprotein IIb/IIIa inhibitors. 8.Antiplatelet activity and oral antithromboticefficacy of L-734,217 J Pharmacol ExpTherap 278, 62–73.

272. Nutt, E.M., Jain, D., Lenny, A.B., Schaffer,L., Siegl, P.K., and Dunwiddie, C.T. (1991)Purification and characterization of recombi-nant antistasin: a leech-derived inhibitor ofcoagulation factor Xa Arch Biochem Biophys285, 37–44.

273. Taniuchi, Y., Sakai, Y., Hisamichi, N.,Kayama, M., Mano, Y., Sato, K., Hirayama,F., Koshio, H., Matsumoto, Y., andKawasaki, T. (1998) Biochemical and phar-macological characterization of YM-60828, a

newly synthesized and orally active inhibitorof human factor Xa Thromb Haemost 79,543–8.

274. Tidwell, R.R., Webster, W.P., Shaver, S.R.,and Geratz, J.D. (1980) Strategies for antico-agulation with synthetic protease inhibitors.Xa inhibitors versus thrombin inhibitorsThromb Res 19, 339–49.

275. Kararli, T.T. (1995) Comparison of thegastrointestinal anatomy, physiology, andbiochemistry of humans and commonly usedlaboratory animals Biopharmaceut Drug Dis-pos 16, 351–80.

276. Sanderson, P.E., Cutrona, K.J., Dorsey, B.D.,Dyer, D.L., McDonough, C.M., Naylor-Olsen, A.M., Chen, I.W., Chen, Z., Cook,J.J., Gardell, S.J., Krueger, J.A., Lewis, S.D.,Lin, J.H., Lucas, B.J., Jr., Lyle, E.A., Lynch,J.J., Jr., Stranieri, M.T., Vastag, K., Shafer,J.A., and Vacca, J.P. (1998) L-374,087,an efficacious, orally bioavailable, pyridinoneacetamide thrombin inhibitor Bioorg MedChem Lett 8, 817–22.

277. Sun, W.Y., Witte, D.P., Degen, J.L., Colbert,M.C., Burkart, M.C., Holmback, K., Xiao,Q., Bugge, T.H., and Degen, S.J. (1998)Prothrombin deficiency results in embryonicand neonatal lethality in mice Proc Natl AcadSci USA 95, 7597–602.

278. Xue, J., Wu, Q., Westfield, L.A., Tuley, E.A.,Lu, D., Zhang, Q., Shim, K., Zheng, X.,and Sadler, J.E. (1998) Incomplete embry-onic lethality and fatal neonatal hemorrhagecaused by prothrombin deficiency in miceProc Natl Acad Sci USA 95, 7603–7.

279. Matsuo, O., Rijken, D.C., and Collen, D.(1981) Thrombolysis by human tissue plas-minogen activator and urokinase in rab-bits with experimental pulmonary embolusNature 291, 590–1.

280. Collen, D., Topol, E.J., Tiefenbrunn, A.J.,Gold, H.K., Weisfeldt, M.L., Sobel, B.E.,Leinbach, R.C., Brinker, J.A., Ludbrook,P.A., and Yasuda, I., et al. (1984) Coro-nary thrombolysis with recombinant humantissue-type plasminogen activator: a prospec-tive, randomized, placebo-controlled trialCirculation 70, 1012–7.

281. Topol, E.J., Califf, R.M., Weisman, H.F.,Ellis, S.G., Tcheng, J.E., Worley, S., Ivanhoe,R., George, B.S., Fintel, D., and Weston, M.,et al. (1994) Randomised trial of coronaryintervention with antibody against plateletIIb/IIIa integrin for reduction of clinicalrestenosis: results at six months. The EPICinvestigators Lancet 343, 881–6.

282. Lynch, J.J., Jr., Sitko, G.R., Lehman, E.D.,and Vlasuk, G.P. (1995) Primary preven-tion of coronary arterial thrombosis with


the factor Xa inhibitor rTAP in a canineelectrolytic injury model Thromb Haemost74, 640–5.

283. (1998) Inhibition of the platelet glyco-protein IIb/IIIa receptor with tirofiban inunstable angina and non-Q-wave myocardialinfarction. Platelet receptor inhibition inischemic syndrome management in patientslimited by unstable signs and symptoms(PRISM-PLUS) study investigators N Engl JMed 338, 1488–97.

284. Ohman, E.M., Kleiman, N.S., Gacioch, G.,Worley, S.J., Navetta, F.I., Talley, J.D.,Anderson, H.V., Ellis, S.G., Cohen, M.D.,Spriggs, D., Miller, M., Kereiakes, D.,Yakubov, S., Kitt, M.M., Sigmon, K.N.,Califf, R.M., Krucoff, M.W., and Topol,E.J. (1997) Combined accelerated tissue-plasminogen activator and platelet glyco-protein IIb/IIIa integrin receptor blockadewith Integrilin in acute myocardial infarction.Results of a randomized, placebo-controlled,dose-ranging trial. IMPACT-AMI investiga-tors Circulation 95, 846–54.

285. Cohen, M., Demers, C., Gurfinkel, E.P.,Turpie, A.G., Fromell, G.J., Goodman,S., Langer, A., Califf, R.M., Fox, K.A.,Premmereur, J., and Bigonzi, F. (1998)Low-molecular-weight heparins in non-ST-segment elevation ischemia: the ESSENCEtrial. Efficacy and safety of subcutaneous

enoxaparin versus intravenous unfractionatedheparin, in non-Q-wave coronary events AmJ Cardiol 82, 19L–24L.

286. Agnelli, G., Pascucci, C., Cosmi, B., andNenci, G.G. (1990) The comparative effectsof recombinant hirudin (CGP 39393) andstandard heparin on thrombus growth in rab-bits Thromb Haemost 63, 204–7.

287. Eriksson, B.I., Wille-Jorgensen, P., Kalebo,P., Mouret, P., Rosencher, N., Bosch, P.,Baur, M., Ekman, S., Bach, D., Lindbratt, S.,and Close, P. (1997) A comparison of recom-binant hirudin with a low-molecular-weightheparin to prevent thromboembolic compli-cations after total hip replacement N Engl JMed 337, 1329–35.

288. Fitzgerald, D.J., Wright, F., and FitzGerald,G.A. (1989) Increased thromboxanebiosynthesis during coronary thrombolysis.Evidence that platelet activation and throm-boxane A2 modulate the response to tissue-type plasminogen activator in vivo Circ Res65, 83–94.

289. Gold, H.K., Torres, F.W., Garabedian, H.D.,Werner, W., Jang, I.K., Khan, A., Hagstrom,J.N., Yasuda, T., Leinbach, R.C., and Newell,J.B., et al. (1993) Evidence for a reboundcoagulation phenomenon after cessation ofa 4-hour infusion of a specific thrombininhibitor in patients with unstable anginapectoris J Am Coll Cardiol 21, 1039–47.

Chapter 3

Heparin and Low-Molecular Weight Heparins in Thrombosisand Beyond

Shaker A. Mousa

Abstract

Heparin and its improved version, low-molecular weight heparin (LMWH), are known to exertpolypharmacological effects at various levels. Early studies focused on the plasma anti-Xa and anti-IIapharmacodynamics of different LMWHs. Other important pharmacodynamic parameters for heparinand LMWH, including effects on vascular tissue factor pathway inhibitor (TFPI) release, inhibition ofinflammation through NFκB, inhibition of key matrix-degrading enzymes, selectin modulation, inhibi-tion of platelet–cancer cell interactions, and inflammatory cell adhesion, help explain the diverse clinicalimpact of this class of agents in thrombosis and beyond.

Key words: Heparin, low-molecular weight heparin, tissue factor pathway inhibitor, angiogenesis,tumor growth, metastasis, asthma, inflammation.

1. Introduction

Heparin was discovered in 1916 by Jay McLean and WilliamHenry Howell, but did not enter clinical trials until 1935(1, 2). It was originally isolated from canine liver cells, henceits name (hepar, Greek for “liver”). Howell isolated a water-soluble polysaccharide anticoagulant which was also termed hep-arin, although it was distinct from the phosphatide preparationspreviously isolated. The first human trials of heparin began in1935, and by 1937 it was clear that Connaught’s heparin wasa safe, readily available, and effective blood anticoagulant (3).

Current commercial heparin is obtained from porcine intes-tine and is produced by basophils and mast cells. Heparin acts


109

110 Mousa

as an anticoagulant, preventing the formation of clots and theextension of existing clots within the blood, as well as acceler-ates thrombolysis when used with standard thrombolytics (4, 5).In recent years, results of clinical studies have clarified the poten-tial and shortcomings of anticoagulant therapy in the preventionand treatment of thromboembolic disorders. The discovery andintroduction of heparin derivatives, such as the low-molecularweight heparins (LMWHs), have increased clinical options forthe management of thromboembolic disorders while enhancingthe safety of therapy. In the United States, LMWHs are currentlyapproved for the prophylaxis and treatment of deep vein throm-bosis (DVT), but the use of LMWH is being expanded for addi-tional indications, such as the management of unstable anginaand non-Q-wave myocardial infarction (6–9). In addition to itsapproved uses, LMWH is currently being tested for several newerindications, namely inflammatory diseases and cancer. Additionalpharmacological studies and well-designed clinical trials in whichmultiple pharmacokinetic and pharmacodynamic parameters areanalyzed will provide much needed data on the unique clinicalprofile of each member of this class of novel agents.

Clinical experience has confirmed the crucial role of heparin/LMWH in the success of platelet glycoprotein (GP)IIb/IIIaantagonists (e.g., abciximab, tirofiban, and eptafibatide) for vari-ous cardiovascular indications, in that these antagonists lack effi-cacy and are associated with increased incidence of myocardialinfarction in the absence of heparin. In addition, clinical trials havedemonstrated clinically relevant effects and improved efficacy ofLMWH as compared to unfractionated heparin (UFH) on thesurvival of cancer patients with DVT. Studies from our laboratoryhave demonstrated a significant role for LMWH and LMWH-releasable tissue factor pathway inhibitor (TFPI) on the regula-tion of angiogenesis, tumor growth, cancer-mediated inflamma-tion, and tumor metastasis. In fact, the anti-angiogenic effectsof LMWH or non-anticoagulant (NA)-LMWH are reversed byanti-TFPI antibodies. Modulation of tissue factor (TF)/VIIanon-coagulant activities by LMWH- or NA-LMWH-releasableTFPI, as well as inhibition of matrix-degrading enzymes andselectins, and the demonstrated anti-inflammatory efficacy ofLMWHs could potentially combine to improve clinical outcomesin patients with vascular thrombosis, cancer, and inflammatorydisorders.

The key reason behind the success of heparin in thrombosisand beyond is its polypharmacological mode of action in prevent-ing and treating diseases that would benefit only slightly from sin-gle pharmacological mechanism-based agents. Thromboembolicdisorders are multi-factorial, driven by hypercoaguable, hyper-active platelet, pro-inflammatory, dysfunctional endothelial, andpro-angiogenic states. Heparin can effectively modulate all

Heparin and Low-Molecular Weight Heparins in Thrombosis and Beyond 111

Table 3.1Molecular targets and polypharmacological effects of LMWHs

Mode/site of action Pharmacological targets

ATIII-dependent plasmatic effects Anti-Xa, Anti-IIa; other coagulation factors

ATIII-independent vascular effects TFPI, NO, vWFModulation of cell adhesion molecules Selectins (P, L, and E); immunoglobulins (soluble

ICAM-1 and VCAM-1)

Fibrinolytic system t-PA, PAI-1Inflammatory mediators TNF-α, IL-6

LMWH, low-molecular weight heparin; ATIII, antithrombin III; TFPI, tissue factor pathway inhibitor; NO, nitricoxide; vWF, vonWillebrand factor; ICAM, inter-cellular adhesion molecule; VCAM, vascular cell adhesion molecule;t-PA, tissue plasminogen activator; PAI-1, plasminogen activator inhibitor-1; TNF, tumor necrosis factor; IL,interleukin.

Table 3.2Potential utility of heparin beyond anticoagulation

Heparin-sensitivedisease state Effects in experimental models Clinical status

Adult respiratorydistress syndrome

Reduces cell activation and accumulation in airways;neutralizes mediators and cytotoxic cell products;improves lung function in animal models

Controlled clinical trials

Allergic rhinitis As for adult respiratory distress syndrome; it has notbeen tested in a specific nasal model


Arthritis Inhibits cell accumulation, collagen destruction, andangiogenesis

Anecdotal reports

Asthma As for adult respiratory distress syndrome; it hasbeen shown to improve lung function in experi-mental models


Cancer Inhibits tumor growth, metastasis, and angiogenesis;increases survival time in animal models

Anecdotal reports plusrecent clinical trials

Inflammatory boweldisease

Inhibits inflammatory cell transport in general; nottested in a specific animal model


of these components, as well as interactions among them(Table 3.1).

Because heparin was discovered over a half century ago,knowledge of the chemical structure and molecular interactionsof this fascinating poly-component was limited in early stagesof development. Through the efforts of major multidisciplinarygroups of researchers and clinicians, it is now well recognized thatheparin has multiple modes/sites of action and can be used formultiple indications (Table 3.2). In the not-too-distant future,we may witness the impact of heparin derivatives on the manage-ment of a variety of diseases.

112 Mousa

2. Heparin VersusLMWH

Heparin is a glycoaminoglycan formed by sulfated oligosaccha-rides. Variation in the number of polymeric units results inheparins of different molecular weights. LMWH is generatedby partial hydrolysis or enzymatic degradation of UFH. In con-trast to UFH, LMWHs have a lower affinity for plasma proteins,endothelial cells, and macrophages (10–12). These differences inbinding characteristics underlie the pharmacokinetic differencesobserved between LMWHs and UFH (Table 3.3). The bind-ing of UFH to plasma proteins reduces its anticoagulant activity,which, combined with variations in the concentration of heparin-binding proteins in plasma, results in an unpredictable anticoag-ulant response.

Table 3.3Clinical profile of UFH versus LMWH

UFH LMWH

Continuous, IV infusion bid or qd subcutaneous injection

Primarily administered in hospital Administered in hospital, office, or homeUsually administered by health-care professionals Administered by patient, caregiver, or professional

Monitoring and dosing adjustments No monitoring; fixed or weight-based dosingFrequent dosing errors More precise dosing

Risk of thrombocytopenia and osteoporosis Decreased risk of adverse eventsInexpensive, but not cost-effective Demonstrated pharmacoeconomic benefits

Requires 5–7 days in the hospital Requires 0–2 days in the hospital

UFH, unfractionated heparin; LMWH, low-molecular weight heparin; IV, intravenous; bid, twice daily; qd, every day

LMWHs exhibit many advantageous properties, includ-ing improved subcutaneous bioavailability, lower protein bind-ing affinity, and longer half-life, as well as less variability inantithrombin (AT)III recognition sites, glycosaminoglycan con-tent, anti-serine protease activity (anti-Xa, anti-IIa, anti-Xa/anti-IIa, and anti-other coagulation factors), induction of TFPIrelease, and vascular endothelial cell (EC) binding kinetics (13–16). For these reasons, over the last decade, LMWHs haveincreasingly replaced UFH for the prevention and treatmentof venous thromboembolic disorders. Randomized clinical trialshave demonstrated that individual LMWHs used at optimizeddoses are at least as effective as and probably safer than UFH.Convenient once- or twice-daily subcutaneous (SC) dosing reg-imens that do not require monitoring have also encouraged the


wide use of LMWHs. It is well established that different LMWHsvary in their physical and chemical properties due to differencesin methods of manufacturing. These differences in turn trans-late into differences in pharmacodynamic and pharmacokineticcharacteristics (17). The World Health Organization (WHO) andUnited States Food and Drug Administration (US FDA) regarddifferent preparations of LMWH as individual drugs that cannotbe used interchangeably (17).

The bioavailability of LMWHs after intravenous (IV) or SCadministration is greater than UFH, ranging from 87 to 98%.The bioavailability of UFH, by contrast, is 15 to 25% after SCadministration. LMWHs have a biological half-life (t1/2

) (basedon anti-Xa clearance) nearly double that of UFH. The t1/2

of theLMWHs enoxaparin, deltaparin, tinzaparin, and others is between100 and 360 min, depending on the route of administration (IVor SC). Anti-Xa activity persists longer than AT activity, whichreflects faster clearance of longer heparin chains (17).

LMWH, in doses based on patient weight, does not requiremonitoring, probably because of the improved bioavailability,longer plasma t1/2

, and more predictable anticoagulant responseinduced by LMWH as compared to UFH when administered SC.Although LMWHs are more expensive than UFH, a pilot studyin pediatric patients found that SC administration of LMWHreduced the number of necessary laboratory assays, nursing hours,and phlebotomy time (18).

LMWHs are expected to continue to supplant the use ofUFH, as programs are developed for new indications and clin-ician comfort with use of these drugs increases. In addition, aspatients and health-care providers recognize the relative simplic-ity of administration by SC injection, together with the real costsavings and quality-of-life benefits associated with reduced hospi-tal stays, the trend toward outpatient use will also continue.

3. EmergingLinks BetweenThrombosisand Inflammation:Potential Roleof Heparin

There are several lines of evidence of the interplay between acti-vated platelets/leukocytes and the coagulation cascade. The real-ization of this dynamic interplay led to the elucidation of a cascadeof events that begins with the exposure of platelet GPIIb/IIIareceptors in the active state, leading to platelet fibrinogen bind-ing and amplification of platelet aggregate formation. Acti-vated platelets also interact with leukocytes, leading to platelet–leukocyte cohesion and leukocyte activation. Hyperactive plateletsprovide a surface for the generation of thrombin, a potent platelet

114 Mousa

and leukocyte activator. In addition, there is significant inter-play between the coagulation cascade, platelets, and the bloodvessel wall during the progression of thromboembolic disorders.Depending on shear level (i.e., venous, or low shear, versus arte-rial, or high shear), platelet/fibrin proportion and contributionsvary.

Infection that leads to the initiation of pro-inflammatorystimuli can be a major predisposing factor in the propagation ofthromboembolic disorders. Endotoxin liberated from Escherichiacoli and other bacteria can induce a pro-inflammatory state, withincreased levels of tumor necrosis factor (TNF)-α and othercytokines (Fig. 3.1). This leads to the activation of leukocytes,increased expression of membrane L-selectin, and the sheddingof soluble L-selectin, which can serve as a surrogate marker ofleukocyte activation. Activation of leukocytes leads to the prop-agation and generation of TF, which initiates and amplifies thehypercoaguable state, as well as the up-regulation of TNF-α pro-duction (Fig. 3.1). The hypercoaguable state and the generationof thrombin activates platelets, leading to the overexpressionof platelet membrane P-selectin and the shedding of solubleP-selectin, which can act as a surrogate marker of platelet activa-tion (19). The pro-inflammatory state can also induce EC insult,leading to increased EC membrane expression and shedding ofsoluble vascular adhesion molecule-1 (VCAM-1), inter-cellularadhesion molecule-1 (ICAM-1), and E-selectin. Emerging linksbetween thrombosis, angiogenesis, and inflammation in vascular,cardiovascular, and inflammatory disorders are shown in Fig. 3.2.A key role of TF/VIIa in angiogenesis, inflammation, and throm-bosis is illustrated in Fig. 3.3.

Pro-inflammatoryStimuli

Natural Anti-inflammatory

factors

Hemostasis Activation(TED)

InflammatoryDiseases

Natural anticoagulants,

Platelet activation inhibitors

Endothelial activation inhibitors

Pro-coagulants

Platelet activation

Endothelial activation

(A)

(B)

Fig. 3.1. Schematic illustration of hemostasis activation in inflammatory diseases and increased inflammation inthromboembolic disorders (TED).


CardiovascularDiseases

Vascular Disorders

Inflammatory Diseases

Fig. 3.2. Emerging links between thrombosis, angiogenesis, and inflammation invascular, cardiovascular, and inflammatory diseases.

FVIIaTF

Possible relationships of TF/VIIa toAngiogenesis, Inflammation & Thrombosis

CoagulationCoagulationand plateletand plateletActivationActivation

ThrombosisThrombosis

ABP-280

InflammationInflammation

AngiogenesisAngiogenesis

VEGF & otherFactors

SignalingMechanism?

InflammationInflammation

Fig. 3.3. TF/VIIa in angiogenesis, inflammation, and thrombosis: Impact of heparinreleasable tissue factor pathway inhibitor (TFPI).

In recent years, several studies have shown that heparin andLMWH have obvious anti-inflammatory activity in addition totheir traditional anticoagulant effects (20, 21). In animal models,heparin disaccharides inhibit TNF-α production by macrophagesand decrease immune inflammation (22). Heparin accelerates thehealing of mucosa in colitis in several clinical studies and has anti-inflammatory effects (23–28). Thus, administration of heparincan induce anti-inflammatory as well as anticoagulant effects.

Heparin is currently used for the treatment and preventionof thrombotic and thromboembolic conditions like DVT, pul-monary embolism (PE), and crescendo angina (29–31). Heparinactivates ATIII to prevent the conversion of fibrinogen to fib-rin and accelerates the inhibition of factors XIIa, XIa, IXa,and Xa. Heparin also possesses non-anticoagulant properties,including the ability to modulate various proteases, as well asanti-complement activity and anti-inflammatory actions. Inhaled

116 Mousa

heparin has been shown to reduce early phase asthmatic reactions(see below) and to suppress allergen-induced rises in bronchialhyper-reactivity. Heparin also inhibits the acute cutaneous reac-tion to allergens. The ubiquitous distribution of heparin in tissuespaces may serve to limit inflammatory responses in tissues whereleukocytes accumulate following an inflammatory challenge. Inthis regard, it is of note that heparin is found in high concen-trations in the gastrointestinal tract and the lung (29, 30), twoorgans that are exposed to the external environment.

The anti-inflammatory activity of heparin has been reinforcedby positive, although small, clinical trials in patients suffering froma range of inflammatory diseases, including rheumatoid arthritisand bronchial asthma (35, 36). In addition, a number of clinicalstudies have recently demonstrated the anti-inflammatory activ-ity of heparin in the treatment of inflammatory bowel disease atdoses that do not produce anti-hemorrhagic complications (36)(see below). It is now well recognized that different portions ofthe heparin molecule mediate anti-inflammatory activity. Giventhat an isolated pentasaccharide sequence retains the ability toinhibit ATIII (37), it is reasonable to expect that we will reachthe point at which the anti-inflammatory and anticoagulant activ-ities of heparin can be separated (38).

3.1. Heparinand Asthma

Heparin significantly reduces asthma symptoms within 10 min ofadministration. This activity of heparin might be related to itsability to prevent the release of histamine from mast cells andinterfere with the stimulation of mast cell mediator secretionthrough blocking internal calcium release. At later time points,there is evidence to suggest that heparin reduces eosinophilrecruitment through several different mechanisms, including pre-vention of mast cell mediator release, or indirectly through down-regulation of adhesion molecules on ECs, thus limiting eosinophilmigration into the nasal mucosa (32–34). There is also evidencethat the heavily anionic character of heparin inactivates platelet-activating factor, a cationic protein with potent chemotactic activ-ity for human eosinophils. Intranasal heparin attenuates the nasalresponse to an allergic challenge in atopic rhinitic subjects, withno adverse reactions (32–34). Additional studies are needed tofully understand the mechanisms by which heparin mediates anti-inflammatory activity and to optimize heparin use in allergic dis-eases such as rhinitis and asthma.

3.2. Heparin/LMWHand InflammatoryBowel Diseases

Several uncontrolled studies have pointed to the potential ther-apeutic benefit of heparin in the clinical management of ulcera-tive colitis and Crohn’s disease. Although these studies includedonly a limited number of patients, they demonstrated appar-ent beneficial effects of heparin with no associated hemorrhagiccomplications (36, 39–45). Heparin has also been shown to


prevent macroscopic inflammatory lesions in an animal modelof experimental colitis. However, the molecular mechanismsby which heparin attenuates disease symptoms in patients withinflammatory bowel disease (IBD) are unknown. The therapeuticactions of heparin observed in IBD patients likely involve atten-uation of inflammatory processes and the hypercoaguable stateassociated with clinical exacerbation of IBD, which could in turnpromote mucosal repair.

3.3. Heparin as anAnti-inflammatoryMolecule: PotentialMechanisms

A large body of evidence supports the concept that heparinhas anti-inflammatory actions. Heparin modulates some of thepathophysiological effects of endotoxin and TNF-α, such as neu-trophil migration, edema formation, pulmonary hypertension,and hypoxemia (20–22, 24, 26, 28). Moreover, heparin has beenshown to suppress specific neutrophil functions, such as superox-ide generation (25) and chemotaxis in vitro (23, 27), as well asreduce eosinophil migration (25) and diminish vascular perme-ability (49, 50). One of the proposed mechanisms by which theanti-inflammatory actions of heparin are mediated is the bind-ing of glycosaminoglycan to adhesion molecules expressed onthe surface of activated ECs and/or leukocytes. Recent in vitrostudies have shown that heparin effectively binds to endothe-lial P-selectin, but not E-selectin (21, 51), and to L-selectin andCD11b/CD18 expressed on neutrophils (52, 53).

Under different experimental and clinical conditions, heparinwas found to actively reduce the process of leukocyte recruit-ment into sites of injury or applied inflammatory stimuli. Salaset al. (46) provided the first in vivo evidence of an anti-migratorymechanism of action of heparin using intravital microscopy. Infact, intravital microscopic techniques have allowed direct obser-vation of inflamed microvascular beds, which helped define theparadigm of white blood cell extravasation. Leukocyte interac-tion with the endothelium of an inflamed post-capillary venule isinitially intermittent and dynamic (cell rolling); it then becomesstatic (firm adhesion) and ultimately culminates with diapedesis(47). Using the potent cytokine TNF-α to promote this cascadeof events in vivo, Salas and colleagues were able to show that hep-arin down-regulated TNF-α-induced leukocyte rolling, adhesion,and migration into gut tissue without affecting changes in vas-cular permeability. These data extended and confirmed previousstudies in which heparin reduced leukocyte adhesion to vascularECs in vitro (48) and recruitment of inflammatory cells into othertissues during experimental inflammatory reactions (38).

There have been some reports of the effect of heparin on reac-tive oxygen species (ROS) generation (54) and cytokine secretion(55) by leukocytes in vitro. It was recently demonstrated that hep-arin, when injected intravenously into normal subjects at a doseof 10,000 IU, inhibits ROS generation by mono-nuclear cells

118 Mousa

(MNC) and polymorphonucleocytes (PMNs) (56). Heparin hasalso been shown in a series of experiments using N-acetyl hep-arin to protect the heart from ischemia-reperfusion injury bothin vivo and in vitro, independently of its antithrombin actions(57, 58). It has been suggested that this protection might be dueto a reduction in complement activation-mediated injury to theheart (57). Since ischemia-reperfusion injury might also be medi-ated by oxidative damage (59), it is possible that the protectiveeffect of heparin could be due to an inhibition of ROS generation.A reduction in superoxide radical formation by heparin is likelyto increase the bioavailability of nitric oxide (NO) for vasodilata-tion. In fact, heparin has been shown to exert a vasodilatory effectin normal subjects in vivo (60). Increased bioavailability of NOcould in turn have the additional beneficial effect of inhibitingleukocyte adhesion to the endothelium, which would inhibit orretard inflammation (61). NO also inhibits the pro-inflammatorytranscription factor NF-κB, which plays a central role in triggeringand coordinating both innate and adaptive immune responses.

4. EmergingLinks BetweenThrombosisand Cancer:Potential Roleof Heparin

The association between activation of the coagulation system andsystemic thrombosis in human cancer has been recognized forover a century, since Trousseau’s original description of migra-tory thrombophlebitis complicating gastrointestinal malignancy(62). Greater appreciation in recent years of the interdependencyof the coagulation system and malignant behavior has led toan understanding of how an activated coagulation system mightenhance cancer cell growth (63). A recent Danish study, while notestablishing causality or even a biologic association, showed thatpatients with cancer who developed venous thrombosis duringthe course of their disease had significantly shorter cancer-relatedsurvival than similar patients who remained thrombosis-free (64).Stronger evidence from several studies, including randomizedclinical trials, has demonstrated improved cancer-related survivalin patients treated with anticoagulants as compared to those notreceiving anticoagulants (65–70).

Thromboembolic events have been shown to be importantpredictors of cancer (69). Thus, cancer screening in patients with-out identifiable risk factors for thrombosis could be helpful forearly detection, diagnosis, and management of disease.

Thrombin generation and fibrin formation can be detectedconsistently in patients with malignancy, who are already atincreased risk of thromboembolic complications. This is impor-tant, because fibrin formation is also involved in the processes


of tumor spread and metastasis. Activation of blood coagulationin cancer is a complex phenomenon, involving many differenthemostatic pathways and numerous interactions of tumor cellswith other blood cells, including platelets, monocytes, and ECs.Tumor cells have the capacity to interact with all components ofthe hemostatic system. They can directly activate the coagulationcascade by producing their own pro-coagulant factors, or theycan stimulate the pro-thrombotic properties of other blood cellcomponents.

The etiology of thrombosis in malignancy is multi-factorial;mechanisms include the release of pro-coagulants by tumorcells, plus other predisposing factors leading to a hypercoaguablestate that is amplified by chemotherapeutic and radio-therapeuticagents (67, 71–75). In fact, unexplained thromboembolismmight be an early indicator of the presence of a malignant tumorbefore signs and symptoms of the tumor itself become obvious.

Hemostatic abnormalities occur in a majority of patientswith metastatic cancer. These abnormalities can be categorizedas follows: increased platelet aggregation and activation; abnor-mal activation of the coagulation cascade; release of plasminogenactivator inhibitor type 1 (PAI-1); and decreased hepatic synthe-sis of anticoagulant proteins like protein C and ATIII. Activationof the coagulation cascade is mediated through the release of TFand other pro-coagulants from the plasma membranes of tumorcells (67, 73).

Increasing evidence suggests that thrombotic episodes mightprecede the diagnosis of cancer by months or years, thus rep-resenting a potential marker for occult malignancy (71). Therehas been recent emphasis on the potential risk of cancer ther-apy (both surgery and chemotherapy) in enhancing the risk forthromboembolic disease (72, 75). Postoperative DVT is indeedmore frequent in patients who have undergone surgery formalignant diseases than for other disorders. Both chemotherapyand hormone therapy are associated with an increased throm-botic risk, which can be prevented by low-dose oral antico-agulation (76, 77). The pro-coagulant activities of tumor cellshave been extensively studied, raising the possibility that a spe-cific tumor pro-coagulant could serve as a novel marker ofmalignancy.

Cancer disturbs those cellular functions that are critical forhomeostasis in multi-cellular organisms, namely, growth, differ-entiation, apoptosis, and tissue integrity. There have been numer-ous clinical and experimental studies showing that invasion resultsfrom cross-talk between cancer cells and host cells (i.e., platelets,myofibroblasts, ECs, and leukocytes, all of which are themselvesinvasive). In bone metastases, for example, host osteoclasts serveas targets for therapy. Molecular analysis of invasion-associatedcellular activities (namely, homotypic and heterotypic cell–cell

120 Mousa

adhesion, cell–matrix interactions and ectopic survival, migration,and proteolysis) has uncovered multiple underlying branched sig-nal transduction pathways and extensive cross-talk between indi-vidual pathways. The role of proteolysis in invasion is not limitedto breakdown of extracellular matrix, but also cleavage of pro-invasive fragments from cell surface GPs.

4.1. Heparinin Platelet–CancerCell Adhesion

Activated platelets release angiogenic growth factors and cantherefore potentially contribute to tumor angiogenesis (78–80).Growth factors released by platelets include vascular endothelialgrowth factor (VEGF), basic fibroblast growth factor (bFGF),and platelet-derived growth factor (PDGF) (78–80). Plateletshave been implicated in tumor biology (81), and serum levels ofVEGF correlate with platelet counts during chemotherapy (82).Furthermore, platelet–tumor cell interactions are believed to beimportant in tumor metastasis. Expression of TF by tumor cellsenhances metastasis and angiogenesis and is primarily respon-sible for tumor-induced thrombin generation and the forma-tion of tumor cell–platelet aggregates. Activated platelets expressand release CD40 ligand (CD40L), which induces endothe-lial TF expression through ligation to CD40. Recent data haveshown that in malignancy, increased cellular TF activity inducedby CD40 (tumor cell)–CD40L (platelet) interactions enhancesintravascular coagulation and hematogenous metastasis (83).Inhibition of experimental metastasis and tumor growth has beendemonstrated in animals by thrombocytopenia and antiplatelettherapies (84, 85).

There are several classic studies demonstrating that the forma-tion of tumor cell–platelet complexes in the bloodstream is impor-tant in facilitating the metastatic process. Metastasis in animalmodels can be inhibited by heparin, and retrospective analyses ofheparin use in human cancer are yielding promising results (87).

4.2. Treatmentof Venousthromboembolism inCancer Patients

A growing body of evidence suggests that a tumor-mediatedhypercoagulation state can develop in cancer patients. There is astrong association between cancer and venous thromboembolism(VTE) (Table 3.4), and patients with cancer are at a remarkablyhigher risk of VTE than patients who are free from malignant dis-orders and who experience prolonged immobilization from anycause or following surgical interventions.

The management of DVT and PE in patients with cancerrepresents a potential dilemma for clinicians. Co-morbidities,warfarin failure, hampered venous access, and a high bleedingrisk are some of the factors that often complicate anticoagulanttherapy in these patients. The use of central venous access devicesis increasing, but optimal treatment of catheter-related throm-bosis remains controversial. UFH is the traditional standard forinitial treatment of VTE, but LMWHs have been shown to be


Table 3.4Mechanisms of induction of a hypercoaguable stateby tumors

Malignant tumors promote

Tumor cell surface TF expressionMacrophage TF expression

Expression of cell surface phospholipids that activate coagulationTumor-mediated platelet activation and accumulation

Tumor-induced endothelial cell activationTumor-mediated monocyte activation

Tumor-mediated fibrin generation

Modified from Mousa (93)TF, tissue factor.

equally safe and effective in hemodynamically stable patients. Forlong-term treatment or secondary prophylaxis, vitamin K antag-onists remain the mainstay of treatment. However, the incon-venience and narrow therapeutic window of oral anticoagulantsmake extended therapy unattractive and problematic. As a result,LMWHs are being evaluated as an alternative for long-term ther-apy (88, 89). The role of inferior vena cava filters in cancerpatients is ill defined, but these devices remain the treatmentof choice in patients with contraindications for anticoagulanttherapy.

In cancer patients affected by DVT, treatment with LMWHhas been reported to lower mortality to a greater extent than stan-dard heparin therapy. Several clinical studies of various LMWHs,including enoxaparin, deltaparin, certoparin, and tinzaparin, havedemonstrated survival benefits as compared to UFH in cancerpatients with certain tumor types and at early stages (90–92).The benefits of LMWH and UFH or warfarin have been exam-ined in ovarian, uterine, lung (small and non-small cell), colorec-tal, and gastric tumors (89–92). The efficacy and safety profile ofLMWH has been shown to be superior to UFH or warfarin (90).This increased efficacy of LMWH as compared to UFH could berelated to the improved pharmacokinetic properties of LMWHand ease of use in and out of the hospital (93). The improvedefficacy and safety of LMWH as compared to warfarin has alsobeen attributed to properties other than its anticoagulant actionsand wider therapeutic index (93). These studies raise the intrigu-ing possibility that LMWHs directly or indirectly modify tumorgrowth progression.

There is convincing evidence for an increased incidence ofnewly diagnosed malignancy among patients with unexplainedVTE during the first 6–12 months after the thromboembolic

122 Mousa

event (62, 94–96). In fact, the existence of a positive feedbackloop between tumor and clot that magnifies each entity has beendemonstrated (93). Tumor fibrin is a consistent feature of tumorstroma and is deposited shortly after tumor cell inoculation (93,97). Since fibrin is likely to be beneficial to tumor growth, it ispossible that the ability of normal or malignant tissue to generatefibrin influences metastasis (97).

4.3. Heparin/LMWH,TFPI,and Anti-neoplasticEffects

Metastasis is the most deleterious event in cancer and leads tomortality in cancer patients (98). Results from animal studies haveshown that metastasis can be inhibited by UFH. There have alsobeen clinical reports of survival benefits from UFH and LMWHthat go beyond their antithrombotic actions. Evidence from sev-eral studies has suggested putative anti-neoplastic mechanisms forheparin and heparin derivatives (Tables 3.5 and 3.6). However,the primary anti-neoplastic mechanisms of heparin remain to befully defined.

Table 3.5Effects of heparin in cancer

Inhibits Stimulates

Angiogenesis Immune system

Proteases Differentiation and apoptosisGrowth factors Nitric oxide production

Coagulation factors TFPI ReleaseOncogene expression

Free radical generationInflammation via NFκB

Modified from Mousa (93)NFκB, nuclear factor κB; TFPI, tissue factor pathway inhibitor.

In an experimental model of injectable B16 metastaticmelanoma, SC injection of tinzaparin before intravenous injec-tion of melanoma cells reduced lung tumor formation inmice (99). Similarly, intravenous injection of TFPI prior totumor cell injection reduced B16 lung metastasis and abolishedtumor cell-induced thrombocytopenia. These results support thepotential role of LMWH and releasable TFPI in tumor growthand metastasis (100). However, it is important to note that hep-arin and LMWH are thought to inhibit tumor metastasis via dif-ferent mechanisms (101, 102).

4.4. Activationof Coagulationin Cancer

Many cancer patients have hemostatic abnormalities that predis-pose them to platelet activation and fibrin formation, leading toclinical or sub-clinical thrombosis (103, 104). Cancer itself leads


Table 3.6Comparison of heparin and heparin-derived compounds

Heparin-derivedcompound Comparison to heparin Biological activity

Heparin tetrasaccharide Non-anticoagulant;non-immunogenic; orally active

Anti-allergy

Pentosan polysulfate Plant derived; minimal anticoagulantactivity; anti-inflammatory; orallyactive

Anti-inflammatory; anti-adhesive;anti-metastatic

Phosphomannopentanosesulfate

Potent inhibitor of heparanase activity Anti-metastatic; anti-angiogenic;anti-inflammatory

Selectively O-desulfatedheparin

Lacks anticoagulant activity Anti-inflammatory; anti-allergy;anti-adhesive

to thrombosis, which in turn, enhances the metastatic spread oftumor cells. Heparin therapy is effective and safe for thrombopro-phylaxis, and LMWH works just as well as or better than UFH.Thus, the antithrombotic action of heparin is a potential mecha-nism by which it can inhibit malignant processes.

TF has been implicated in the up-regulation of pro-angiogenic factors, such as VEGF, by tumor cells. The mechanisminvolves complex interactions between tumor cells, macrophages,and ECs, leading to TF expression, fibrin formation, and tumorangiogenesis (105). A recent study has suggested that thrombingeneration occurs via the extrinsic (TF-dependent) coagulationpathway on cell surfaces and that some chemotherapeutic agentsare able to up-regulate TF mRNA and protein expression in can-cer cells (106).

Activation of the blood coagulation system stimulates thegrowth and dissemination of cancer cells through multiple mech-anisms. Laboratory data on the effects of anticoagulants on vari-ous tumors suggest that this treatment approach has considerablepotential in some cancers but not others. For example, renal cellcarcinoma (RCC) is one of a small number of human tumors inwhich the tumor cells contain an intact coagulation pathway, lead-ing to thrombin generation and conversion of fibrinogen to fibrinimmediately adjacent to viable tumor cells (107). This is also truefor melanoma, ovarian cancer, and small cell lung cancer, but notbreast, colorectal, or non-small cell lung cancers (108). This is ofconsiderable relevance given that the growth of melanoma andsmall cell lung cancer is inhibited by anticoagulants, but not thelatter three tumor types (65). Based on the unique features ofthe interaction of RCC with the coagulation system, RCC mightalso respond to anticoagulation therapy in a similar manner assmall cell lung cancer and melanoma. Thus, anticoagulants thatact at the level of TF/VIIa level might have improved efficacy

124 Mousa

and safety in inhibiting tumor-associated thrombosis, angiogene-sis, and metastasis. See Table 3.6 for a list of the molecular targetsand polypharmacological effects of LMWHs.

Tumors differ in the nature of their interactions with thecoagulation system. In this regard there are two main types oftumors: (1) those that activate the coagulation system directlyand (2) those that mediate coagulation activation indirectly viaa paracrine mechanism. Tumors in the first group include RCC,melanoma, and ovarian and small cell lung cancers. These tumorsoverexpress pro-coagulant molecules such as TF and cancer pro-coagulant. The entire coagulation pathway is assembled on thetumor cell surface, leading to fibrin formation in close prox-imity to the tumors. This explains in part the occasional find-ing in RCC of clots emanating from the tumor and extendinginto the renal vein and inferior vena cava. Tumors in the secondgroup tend to activate systemic coagulation through the releaseof cytokines [i.e., TNF-α, interleukin (IL)-1β] that in turn stim-ulate the production of pro-coagulant molecules on the surfaceof circulating monocytes. Examples of these tumor types includebreast, colorectal, and non-small cell lung cancers. Based on thesedifferences in mechanism of coagulation activation, it is reason-able to predict that tumors in the first group would be morelikely to respond to anticoagulants that interfere with TF/VIIathan tumors in the second group. In support of this hypothe-sis, prospective trials have shown that anticoagulants are stronglyactive in melanoma and small cell lung cancer, but not in breast,colorectal, and non-small cell lung cancers (65, 66, 68, 70, 109).

4.5. CombinationAnticoagulantand AntiplateletTherapies in Cancer

The processes of blood coagulation and generation of new bloodvessels play crucial roles in wound healing. Platelets, for exam-ple, are the first line of defense during vascular injury and con-tain at least a dozen promoters of angiogenesis, the secretionof which into the surrounding vasculature can be induced uponplatelet activation by thrombin (85). It follows that these pro-cesses are also intricately linked within human tumors. Targetingboth the coagulation and angiogenesis pathways could result inmore potent anti-tumor effects than targeting either pathwayalone. Furthermore, elucidation of the TF signaling pathway intumor cells should provide new insight into the normal cellu-lar biology of TF and TF-mediated signaling in ECs, smoothmuscle cells, and fibroblasts. New classes of anticoagulantmolecules have recently been developed that selectively target TFand/or the TF/VIIa complex (86, 87, 110), opening the door toa better understanding of this pathway, and providing a rationalbasis for the development of new agents to prevent and/or reduceangiogenesis-related disorders and tumor-associated thrombosisand modulate the positive feedback loop between thrombosis andcancer (111).


Based on the key role of fibrin and platelets in tumor sur-vival, angiogenesis, and metastasis (106, 112), several groups havedemonstrated a synergistic benefit of combined anticoagulant andantiplatelet therapy at adjusted doses (113).

5. Heparinand Angiogenesis

As discussed above, the coagulation system is activated in mostcancer patients and plays an important role in tumor biology. Itcan also contribute substantially to tumor angiogenesis, a pro-cess that represents an imbalance in the normal mechanismsthat allow organized healing after injury. A steadily growing andincreasingly appreciated body of knowledge of the relationshipbetween the coagulation and the angiogenesis systems has impor-tant research and clinical implications. Manipulation of these sys-tems could potentially minimize neo-angiogenesis, which is essen-tial for tumor growth, and cancer-associated thromboemboliccomplications.

Angiogenesis is a process that is dependent on the coor-dinated production of stimulatory (angiogenic) and inhibitory(angiostatic) molecules. Any imbalance in the regulatory circuitthat governs this process can potentially lead to the developmentof a number of angiogenesis-mediated diseases. Angiogenesis is amultistep process of forming new vessels through sprouting frompreexisting vessels. It involves activation, adhesion, migration,proliferation, and transmigration of ECs across cell matrices toor from new capillaries and from existing vessels. The combineddefects of overproduction of positive regulators of angiogenesisand deficiencies in endogenous angiostatic mediators has beendocumented in tumor angiogenesis, psoriasis, rheumatoid arthri-tis, and other neovascularization-mediated disorders (114, 115).

In a several experimental studies, LMWH exhibits potentanti-angiogenesis effects, inhibition of tumor growth, andsuppression of metastasis (49, 99, 116, 117). In addition,heparin-sensitive endothelial P-selectin has been shown to beinvolved in metastasis (118).

TF, FGF2, VEGF, and IL-8 are pro-angiogenic molecules(105) that can be inhibited by heparin. However, the mechanismsby which heparin counteracts the functions of these factors dif-fer. The natural inhibitor of TF is TFPI. In the presence of hep-arin, Zhang et al. (119) showed that TFPI activity is enhanced,and the stimulatory effects of TF on angiogenesis are reduced.Chemokines such as IL-8, on the other hand, have positivelycharged domains (120). The interaction of heparin with thesepositive domains could mediate the inhibitory effects of heparinon IL-8.

126 Mousa

In addition to angiogenesis, the adhesion of cells in areas awayfrom the primary tumor growth is a key component of metasta-sis. Selectins and integrins are families of cellular receptors thatmediate cell adhesion and are involved in triggering complex sig-naling cascades following EC activation. Heparin inhibits selectin-and integrin-mediated interactions with tumor cells (121), whichsuggests that tumor cells can function as ligands and activate thesecellular pathways.

The LMWH tinzaparin, anti-VIIa antibody (anti-VIIa), andrecombinant TFPI (r-TFPI) have been shown to modulateangiogenesis-related processes, including in vitro endothelial tubeformation and in vivo angiogenesis mediated by angiogenic fac-tors or by cancer cells (117). Tinzaparin, anti-VIIa, and r-TFPIsignificantly inhibit EC tube formation at comparable levels ina concentration-dependent manner, and all three agents blockFGF2-induced angiogenesis in the chick chorioallantoic mem-brane (CAM) model. Furthermore, significant inhibition of colonor lung carcinoma-induced angiogenesis and tumor growth andtumor regression have been demonstrated with tinzaparin, anti-VIIa, and r-TFPI (77). These studies strongly support a signifi-cant role for tinzaparin, anti-VIIa, and tinzaparin-releasable TFPIin the regulation of angiogenesis and tumor growth (117).

6. Conclusions

Available data from uncontrolled studies suggest heparin iseffective in steroid-resistant ulcerative colitis, with 70% completeclinical remission after an average of 4–6 weeks of therapy. Theadministration of heparin for this indication, however, is not cur-rently justified by the limited available data. LMWH was used ina single trial in patients with steroid-refractory ulcerative colitis,and the results were similar to that of heparin. Given that a pro-thrombotic state has been described in IBD, and microvascularintestinal occlusion seems to play a role in the pathogenesis ofIBD, it is reasonable to suggest that the beneficial effects of hep-arin in IBD are due in part to its anticoagulant properties. Beyondits well-known anticoagulant activity, heparin also exhibits a broadspectrum of immune-modulatory and anti-inflammatory proper-ties, through inhibition of neutrophil recruitment and reductionof pro-inflammatory cytokines. Heparin and heparin derivativesrepresent a safe therapeutic option for severe, steroid-resistantulcerative colitis and other inflammatory disorders, althoughrandomized, controlled trials are needed to confirm theseconclusions.


Many cancer patients reportedly develop a hypercoaguablestate with recurrent thrombosis due to the impact of cancer cells,chemotherapy, radiation, immobility, and catheters on activationof the coagulation cascade. Experimental studies have demon-strated that UFH or LMWH interferes with various processes thatregulate tumor growth and metastasis, but these results still needto be clinically validated. These processes might include fibrin for-mation, binding of heparin to angiogenic growth factors such asFGF2 and VEGF, modulation of TF, TFPI release, inhibition ofmatrix-degrading enzymes, and others. Clinical trials demonstratea more clinically relevant effect of LMWH as compared to UFHon the survival of cancer patients with DVT, a finding that needsto be validated in a large multi-center trial of cancer patients withtumors of defined type and stage. Recent results have defined therole of LMWH, anti-factor VIIa, and r-TFPI in the modulationof angiogenesis, tumor growth, and tumor metastasis. There isalso accumulating evidence that antiplatelet drugs could provideadditional benefit in reducing tumor metastasis.

7. FuturePerspective

The next 5–10 years should see the use of anticoagulants with orwithout antiplatelet agents such as aspirin becoming more com-mon as a supplement in cancer patients and other patients with ahigh risk of thrombosis. In addition, heparin derivatives will likelyprove to be more effective than standard single mechanism-basedagents for cancer prevention and treatment.

References

1. Linhardt, R. (1991) Heparin: an importantdrug enters its seventh decade Chem Indust2, 45–50.

2. Marcum, J.A. (2000) The origin of the dis-pute over the discovery of heparin J Hist MedAllied Sci 55, 37–66.

3. Ruty, C. Miracle Blood Lubricant: Con-naught and the Story of Heparin,1928–1937 Health Heritage ResearchServices, www.healthheritageresearch.com.Originally published in CONNTACT (1996)9. Accessed on May 5, 2007.

4. Bjork, I., Lindahl, U. (1982) Mechanism ofthe anticoagulant action of heparin Mol CellBiochem 48, 161–82.

5. Hirsh, J., and Raschke, R. (2004) Hep-arin and low-molecular-weight heparin: theseventh ACCP conference on antithrom-

botic and thrombolytic therapy Chest 126,188S–203S.

6. Aguilar, D., and Goldhaber, S.Z. (1999)Clinical uses of low-molecular-weight hep-arins Chest 115, 1418–23.

7. Mousa, S., Fareed, J. (2001) Advances inanticoagulant, antithrombotic and throm-bolytic drugs Expert Opin Invest Drugs 10,157–62.

8. Mousa, S.A. (2000) Comparative efficacyof different low-molecular-weight hep-arins (LMWHs) and drug interactions withLMWH: implications for management ofvascular disorders Semin Thromb Hemost26(Suppl 1), 39–46.

9. Weitz, J.I. (1997) Low-molecular-weight heparins N Engl J Med 337,688–98.

128 Mousa

10. Hirsh, J., and Levine, M.N. (1992) Lowmolecular weight heparin Blood 79, 1–17.

11. Linhardt, R.J.,and Gunay, N.S. (1999)Production and chemical processing of lowmolecular weight heparins Semin ThrombHemost 25(Suppl 3), 5–16.

12. Nielsen, J.I., and Ostergaard, P. (1988)Chemistry of heparin and low molecularweight heparin Acta Chir Scand Suppl 543,52–6.

13. Brieger, D., and Dawes, J. (1997) Produc-tion method affects the pharmacokinetic andex vivo biological properties of low molec-ular weight heparins Thromb Haemost 77,317–22.

14. Emanuele, R.M., and Fareed, J. (1987) Theeffect of molecular weight on the bioavail-ability of heparin Thromb Res 48, 591–6.

15. Mousa, S.A., Fareed, J., Iqbal, O., andKaiser, B. (2004) Tissue factor pathwayinhibitor in thrombosis and beyond MethodsMol Med 93, 133–55.

16. Mousa, S.A., and Mohamed, S. (2004) Inhi-bition of endothelial cell tube formation bythe low molecular weight heparin, tinzaparin,is mediated by tissue factor pathway inhibitorThromb Haemost 92, 627–33.

17. Boneu, B., Caranobe, C., Cadroy, Y., Dol, F.,Gabaig, A.M., Dupouy, D., and Sie, P.(1988) Pharmacokinetic studies of standardunfractionated heparin, and low molecularweight heparins in the rabbit Semin ThrombHemost 14, 18–27.

18. Sutor, A.H., Massicotte, P., Leaker, M.,and Andrew, M. (1997) Heparin therapy inpediatric patients Semin Thromb Hemost 23,303–19.

19. Taheri, S.A., Lazar, L., Haddad, G.,Castaldo, R., Wilson, M., and Mousa, S.(1998) Diagnosis of deep venous thrombo-sis by use of soluble necrosis factor receptorAngiology 49, 537–41.

20. Lantz, M., Thysell, H., Nilsson, E., andOlsson, I. (1991) On the binding of tumornecrosis factor (TNF) to heparin and therelease in vivo of the TNF-binding protein Iby heparin J Clin Invest 88, 2026–31.

21. Nelson, R.M., Cecconi, O., Roberts, W.G.,Aruffo, A., Linhardt, R.J., and Bevilacqua,M.P. (1993) Heparin oligosaccharides bindL- and P-selectin and inhibit acute inflamma-tion Blood 82, 3253–8.

22. Tyrell, D.J., Kilfeather, S., and Page, C.P.(1995) Therapeutic uses of heparin beyondits traditional role as an anticoagulant TrendsPharmacol Sci 16, 198–204.

23. Bazzoni, G., Beltran Nunez, A., Mascellani,G., Bianchini, P., Dejana, E., andDel Maschio, A. (1993) Effect of hep-

arin, dermatan sulfate, and related oligo-derivatives on human polymorphonuclearleukocyte functions J Lab Clin Med 121,268–75.

24. Darien, B.J., Fareed, J., Centgraf, K.S., Hart,A.P., MacWilliams, P.S., Clayton, M.K.,Wolf, H., and Kruse-Elliott, K.T. (1998)Low molecular weight heparin prevents thepulmonary hemodynamic and pathomorpho-logic effects of endotoxin in a porcine acutelung injury model Shock 9, 274–81.

25. Hiebert, L.M., and Liu, J.M. (1990) Heparinprotects cultured arterial endothelial cellsfrom damage by toxic oxygen metabolitesAtherosclerosis 83, 47–51.

26. Hocking, D., Ferro, T.J., and Johnson, A.(1992) Dextran sulfate and heparin sulfateinhibit platelet-activating factor-induced pul-monary edema J Appl Physiol 72, 179–85.

27. Matzner, Y., Marx, G., Drexler, R., andEldor, A. (1984) The inhibitory effect ofheparin and related glycosaminoglycans onneutrophil chemotaxis Thromb Haemost 52,134–7.

28. Meyer, J., Cox, C.S., Herndon, D.N.,Nakazawa, H., Lentz, C.W., Traber, L.D.,and Traber, D.L. (1993) Heparin in exper-imental hyperdynamic sepsis Crit Care Med21, 84–9.

29. Majerus, P., Broze, J., and Miletich, J.Anticoagulant, thrombolytic and antiplateletdrugs. In: Hardman, J., Limbird, L.,Molinoff, P., Ruddon, R., and Goodman,A., eds. Goodman & Gilman’s, The Phar-macological Basis of Therapeutics. New York:McGraw-Hill;1996:1341–59.

30. Silverstein, R. Drugs affecting hemostosis.In: Munson, P., Mueller, R., and Breeze,G., eds. Munson’s Principles of Pharma-cology. New York: Chapman & Hall;1995:1123–43.

31. Theroux, P., Waters, D., Qiu, S., McCans,J., de Guise, P., and Juneau, M. (1993)Aspirin versus heparin to prevent myocardialinfarction during the acute phase of unstableangina Circulation 88, 2045–8.

32. Diamant, Z., Timmers, M.C., van der Veen,H., Page, C.P., van der Meer, F.J., andSterk, P.J. (1996) Effect of inhaled heparinon allergen-induced early and late asthmaticresponses in patients with atopic asthma AmJ Respir Crit Care Med 153, 1790–5.

33. Garrigo, J., Danta, I., and Ahmed, T. (1996)Time course of the protective effect ofinhaled heparin on exercise-induced asthmaAm J Respir Crit Care Med 153, 1702–7.

34. Vancheri, C., Mastruzzo, C., Armato, F.,Tomaselli, V., Magri, S., Pistorio, M.P.,LaMicela, M., D’Amico, L., and Crimi, N.


(2001) Intranasal heparin reduces eosinophilrecruitment after nasal allergen challenge inpatients with allergic rhinitis J Allergy ClinImmunol 108, 703–8.

35. Bowler, S.D., Smith, S.M., and Lavercombe,P.S. (1993) Heparin inhibits the immediateresponse to antigen in the skin and lungsof allergic subjects Am Rev Respir Dis 147,160–3.

36. Dwarakanath, A.D., Yu, L.G., Brookes,C., Pryce, D., and Rhodes, J.M. (1995)‘Sticky’ neutrophils, pathergic arthritis, andresponse to heparin in pyoderma gangreno-sum complicating ulcerative colitis Gut 37,585–8.

37. Petitou, M., Herault, J.P., Bernat, A.,Driguez, P.A., Duchaussoy, P., Lormeau,J.C., and Herbert, J.M. (1999) Synthesis ofthrombin-inhibiting heparin mimetics with-out side effects Nature 398, 417–22.

38. Tyrrell, D.J., Horne, A.P., Holme, K.R.,Preuss, J.M., and Page, C.P. (1999) Heparinin inflammation: potential therapeutic appli-cations beyond anticoagulation Adv Pharma-col 46, 151–208.

39. Brazier, F., Yzet, T., Boruchowicz, A.,Colombel, J.F., Duckmann, J.C., and Dupas,J.L. (1996) Treatment of ulcerative colitiswith heparin Gastroenterol 110, A872.

40. Brazier, F., T, Y., Duchmann, J., Igliki, F.,and Dupas, J. (1996) Effect of heparin treat-ment on extraintestinal manifestations asso-ciated with inammatory bowel disease Gas-troenterology 110, A872.

41. Dupas, J., Brazier, F., Yzet, T., Roussel, V.,Duchmann, J., and Iglicki, F. (1996) Treat-ment of active Crohn’s disease with heparinGastroenterology 110.

42. Evans, R., and Rhodes, J.M. (1995) Treat-ment of corticosteroid-resistant ulcerativecolitis with heparin: a report of nine cases Gut37, A49.

43. Folwaczny, C., Spannagl, M., Wiebecke, W.,Jochum, M., Heldwein, W., and Loeschke,K. (1996) Heparin in the treatment of highlyactive inflammatory bowel disease (IBD)Gastroenterology 110.

44. Gaffney, P., and Gaffney, A. (1996) Heparintherapy in refractory ulcerative colitis – anupdate Gastroenterology 110, A913.

45. Gaffney, P.R., Doyle, C.T., Gaffney, A.,Hogan, J., Hayes, D.P., and Annis, P.(1995) Paradoxical response to heparin in 10patients with ulcerative colitis Am J Gastroen-terol 90, 220–3.

46. Salas, A., Sans, M., Soriano, A., Reverter,J.C., Anderson, D.C., Pique, J.M., andPanes, J. (2000) Heparin attenuates TNF-alpha induced inflammatory response

through a CD11b dependent mechanismGut 47, 88–96.

47. Panes, J., Perry, M., and Granger, D.N.(1999) Leukocyte-endothelial cell adhesion:avenues for therapeutic intervention Br JPharmacol 126, 537–50.

48. Lever, R., Hoult, J.R., and Page, C.P. (2000)The effects of heparin and related moleculesupon the adhesion of human polymorphonu-clear leucocytes to vascular endothelium invitro Br J Pharmacol 129, 533–40.

49. Carr, J. (1979) The anti-inflammatory actionof heparin: heparin as an antagonist to his-tamine, bradykinin and prostaglandin E1Thromb Res 16, 507–16.

50. Teixeira, M.M., and Hellewell, P.G. (1993)Suppression by intradermal administrationof heparin of eosinophil accumulation butnot oedema formation in inflammatory reac-tions in guinea-pig skin Br J Pharmacol 110,1496–500.

51. Koenig, A., Norgard-Sumnicht, K., Linhardt,R., and Varki, A. (1998) Differential inter-actions of heparin and heparan sulfate gly-cosaminoglycans with the selectins. Impli-cations for the use of unfractionated andlow molecular weight heparins as therapeuticagents J Clin Invest 101, 877–89.

52. Benimetskaya, L., Loike, J.D., Khaled, Z.,Loike, G., Silverstein, S.C., Cao, L.,el Khoury, J., Cai, T.Q., and Stein, C.A.(1997) Mac-1 (CD11b/CD18) is anoligodeoxynucleotide-binding protein NatMed 3, 414–20.

53. Diamond, M.S., Alon, R., Parkos, C.A.,Quinn, M.T., and Springer, T.A. (1995)Heparin is an adhesive ligand for the leuko-cyte integrin Mac-1 (CD11b/CD1) J CellBiol 130, 1473–82.

54. Itoh, K., Nakao, A., Kishimoto, W., andTakagi, H. (1995) Heparin effects on super-oxide production by neutrophils Eur SurgRes 27, 184–8.

55. Attanasio, M., Gori, A.M., Giusti, B., Pepe,G., Comeglio, P., Brunelli, T., Prisco, D.,Abbate, R., Gensini, G.F., and Neri Serneri,G.G. (1998) Cytokine gene expressionin human LPS- and IFNgamma-stimulatedmononuclear cells is inhibited by heparinThromb Haemost 79, 959–62.

56. Dandona, P., Qutob, T., Hamouda, W.,Bakri, F., Aljada, A., and Kumbkarni, Y.(1999) Heparin inhibits reactive oxygenspecies generation by polymorphonuclearand mononuclear leucocytes Thromb Res 96,437–43.

57. Black, S.C., Gralinski, M.R., Friedrichs, G.S.,Kilgore, K.S., Driscoll, E.M., and Lucch-esi, B.R. (1995) Cardioprotective effects of

130 Mousa

heparin or N-acetylheparin in an in vivomodel of myocardial ischaemic and reperfu-sion injury Cardiovasc Res 29, 629–36.

58. Friedrichs, G.S., Kilgore, K.S., Manley, P.J.,Gralinski, M.R., and Lucchesi, B.R. (1994)Effects of heparin and N-acetyl heparin onischemia/reperfusion-induced alterations inmyocardial function in the rabbit isolatedheart Circ Res 75, 701–10.

59. Lucchesi, B.R., and Kilgore, K.S. (1997)Complement inhibitors in myocardialischemia/reperfusion injury Immunophar-macology 38, 27–42.

60. Hawari, F.I., Shykoff, B.E., and Izzo, J.L., Jr.(1998) Heparin attenuates norepinephrine-induced venoconstriction Vasc Med 3,95–100.

61. De Caterina, R., Libby, P., Peng, H.B.,Thannickal, V.J., Rajavashisth, T.B.,Gimbrone, M.A., Jr., Shin, W.S., and Liao,J.K. (1995) Nitric oxide decreases cytokine-induced endothelial activation. Nitric oxideselectively reduces endothelial expression ofadhesion molecules and proinflammatorycytokines J Clin Invest 96, 60–8.

62. Trousseau, A. Phlegmasia alba dolens. In:Baillier, J., ed. Clinique de l’Hotel-Dieu deParis, 2nd edition. London: New SydenhamSociety;1865:654–712.

63. Schiller, H., Bartscht, T., Arlt, A., Zahn,M.O., Seifert, A., Bruhn, T., Bruhn, H.D.,and Gieseler, F. (2002) Thrombin as asurvival factor for cancer cells: thrombin acti-vation in malignant effusions in vivo andinhibition of idarubicin-induced cell death invitro Int J Clin Pharmacol Ther 40, 329–35.

64. Sorensen, H.T., Mellemkjaer, L., Olsen,J.H., and Baron, J.A. (2000) Prognosis ofcancers associated with venous thromboem-bolism N Engl J Med 343, 1846–50.

65. Caine, G.J., and Lip, G.Y. (2002) Anticoag-ulants as anticancer agents? Lancet Oncol 3,591–2.

66. Chahinian, A.P., Propert, K.J., Ware, J.H.,Zimmer, B., Perry, M.C., Hirsh, V., Skarin,A., Kopel, S., Holland, J.F., and Comis, R.L.,et al. (1989) A randomized trial of anti-coagulation with warfarin and of alternat-ing chemotherapy in extensive small-cell lungcancer by the cancer and leukemia group BJ Clin Oncol 7, 993–1002.

67. Goldberg, R.J., Seneff, M., Gore, J.M.,Anderson, F.A., Jr., Greene, H.L., Wheeler,H.B., and Dalen, J.E. (1987) Occult malig-nant neoplasm in patients with deep venousthrombosis Arch Intern Med 147, 251–3.

68. Lebeau, B., Chastang, C., Brechot, J.M.,Capron, F., Dautzenberg, B., Delaisements,C., Mornet, M., Brun, J., Hurdebourcq, J.P.,

and Lemarie, E. (1994) Subcutaneous hep-arin treatment increases survival in small celllung cancer. “Petites Cellules” group Cancer74, 38–45.

69. Saba, H.I., Khalil, F.K., Morelli, G.A., andFoulis, P.R. (2003) Thromboembolism pre-ceding cancer: a correlation study Am JHematol 72, 109–14.

70. Thornes, R. Coumarins, melanoma andcellular immunity. In: McBrien, D., andSlator, T., eds. Protective Agents in Cancer.London: Academic Press;1983:43–56.

71. Baron, J.A., Gridley, G., Weiderpass, E.,Nyren, O., and Linet, M. (1998) Venousthromboembolism and cancer Lancet 351,1077–80.

72. Falanga, A. (1998) Mechanisms of hyper-coagulation in malignancy and duringchemotherapy Haemostasis 28 (Suppl 3),50–60.

73. Kakkar, A.K., DeRuvo, N., Chinswang-watanakul, V., Tebbutt, S., and Williamson,R.C. (1995) Extrinsic-pathway activation incancer with high factor VIIa and tissue factorLancet 346, 1004–5.

74. Levine, M.N. (1997) Prevention of throm-botic disorders in cancer patients undergoingchemotherapy Thromb Haemost 78, 133–6.

75. Rickles, F.R., and Edwards, R.L. (1983)Activation of blood coagulation in cancer:Trousseau’s syndrome revisited Blood 62,14–31.

76. Koopman, M.M., Prandoni, P., Piovella, F.,Ockelford, P.A., Brandjes, D.P., van derMeer, J., Gallus, A.S., Simonneau, G.,Chesterman, C.H., and Prins, M.H. (1996)Treatment of venous thrombosis with intra-venous unfractionated heparin administeredin the hospital as compared with subcuta-neous low-molecular-weight heparin admin-istered at home. The Tasman study group NEngl J Med 334, 682–7.

77. Simonneau, G., Sors, H., Charbonnier, B.,Page, Y., Laaban, J.P., Azarian, R., Laurent,M., Hirsch, J.L., Ferrari, E., Bosson, J.L.,Mottier, D., and Beau, B. (1997) A compar-ison of low-molecular-weight heparin withunfractionated heparin for acute pulmonaryembolism. The THESEE study group.Tinzaparine ou heparine standard: evalua-tions dans l’Embolie pulmonaire N Engl JMed 337, 663–9.

78. Manegold, P.C., Hutter, J., Pahernik, S.A.,Messmer, K., and Dellian, M. (2003)Platelet-endothelial interaction in tumorangiogenesis and microcirculation Blood101, 1970–6.

79. Pintucci, G., Froum, S., Pinnell, J., Mignatti,P., Rafii, S., and Green, D. (2002) Trophic


effects of platelets on cultured endothe-lial cells are mediated by platelet-associatedfibroblast growth factor-2 (FGF-2) and vas-cular endothelial growth factor (VEGF)Thromb Haemost 88, 834–42.

80. Werther, K., Christensen, I.J., and Nielsen,H.J. (2002) Determination of vascularendothelial growth factor (VEGF) in circu-lating blood: significance of VEGF in vari-ous leucocytes and platelets Scand J Clin LabInvest 62, 343–50.

81. Verheul, H.M., and Pinedo, H.M. (1998)Tumor growth: a putative role for platelets?Oncologist 3, i–0i.

82. Verheul, H.M., Hoekman, K., Luykx-deBakker, S., Eekman, C.A., Folman, C.C.,Broxterman, H.J., and Pinedo, H.M. (1997)Platelet: transporter of vascular endothelialgrowth factor Clin Cancer Res 3, 2187–90.

83. Amirkhosravi, A., Amaya, M., Desai, H.,and Francis, J.L. (2002) Platelet-CD40 lig-and interaction with melanoma cell andmonocyte CD40 enhances cellular procoag-ulant activity Blood Coagul Fibrinolysis 13,505–12.

84. Gasic, G.J., Gasic, T.B., and Stewart, C.C.(1968) Antimetastatic effects associated withplatelet reduction Proc Natl Acad Sci USA61, 46–52.

85. Karpatkin, S., Pearlstein, E., Ambrogio, C.,and Coller, B.S. (1988) Role of adhesive pro-teins in platelet tumor interaction in vitro andmetastasis formation in vivo J Clin Invest 81,1012–9.

86. Trikha, M., Zhou, Z., Timar, J., Raso, E.,Kennel, M., Emmell, E., and Nakada, M.T.(2002) Multiple roles for platelet GPIIb/IIIaand alphavbeta3 integrins in tumor growth,angiogenesis, and metastasis Cancer Res 62,2824–33.

87. Varki, N.M., and Varki, A. (2002) Heparininhibition of selectin-mediated interactionsduring the hematogenous phase of carcinomametastasis: rationale for clinical studies inhumans Semin Thromb Hemost 28, 53–66.

88. Kakkar, A.K., and Williamson, R.C. (1997)Prevention of venous thromboembolism incancer using low-molecular-weight heparinsHaemostasis 27 (Suppl 1), 32–7.

89. Zacharski, L.R., and Ornstein, D.L. (1998)Heparin and cancer Thromb Haemost 80,10–23.

90. Meyer, G., Marjanovic, Z., Valcke, J.,Lorcerie, B., Gruel, Y., Solal-Celigny, P.,Le Maignan, C., Extra, J.M., Cottu, P.,and Farge, D. (2002) Comparison of low-molecular-weight heparin and warfarin forthe secondary prevention of venous throm-boembolism in patients with cancer: a ran-

domized controlled study Arch Intern Med162, 1729–35.

91. von Tempelhoff, G.F., Harenberg, J.,Niemann, F., Hommel, G., Kirkpatrick, C.J.,and Heilmann, L. (2000) Effect of lowmolecular weight heparin (Certoparin) ver-sus unfractionated heparin on cancer survivalfollowing breast and pelvic cancer surgery:a prospective randomized double-blind trialInt J Oncol 16, 815–24.

92. Zacharski, L.R. (2002) Anticoagulants incancer treatment: malignancy as a solid phasecoagulopathy Cancer Lett 186, 1–9.

93. Mousa, S.A. (2002) Anticoagulants inthrombosis and cancer: the missing linkSemin Thromb Hemost 28, 45–52.

94. Dvorak, H.F. (1987) Thrombosis and cancerHum Pathol 18, 275–84.

95. Haward, W. (1906) Phlebitis and thrombosisLancet 1, 650–5.

96. Kakkar, A.K., Levine, M.N., Prandoni, P.,Lee, A.Y., Rasmussen, M.S., and Kuter, D.(2002) What is the role for antithromboticsin cancer care? Interactive session with paneldiscussion Cancer Treat Rev 28, 157–9.

97. Dvorak, H.F., Senger, D.R., and Dvorak,A.M. (1983) Fibrin as a component of thetumor stroma: origins and biological signifi-cance Cancer Metastasis Rev 2, 41–73.

98. Chambers, A.F. (1999) The metastatic pro-cess: basic research and clinical implicationsOncol Res 11, 161–8.

99. Amirkhosravi, A., Mousa, S.A., Amaya, M.,and Francis, J.L. (2003) Antimetastatic effectof tinzaparin, a low-molecular-weight hep-arin J Thromb Haemost 1, 1972–6.

100. Mousa, S.A. (2002) Anticoagulants inthrombosis and cancer: the missing linkExpert Rev Anticancer Ther 2, 227–33.

101. Engelberg, H. (1999) Actions of heparin thatmay affect the malignant process Cancer 85,257–72.

102. Varki, A., and Varki, N.M. (2001) P-selectin,carcinoma metastasis and heparin: novelmechanistic connections with therapeuticimplications Braz J Med Biol Res 34,711–7.

103. Falanga, A., and Rickles, F.R. (1999) Patho-physiology of the thrombophilic state in thecancer patient Semin Thromb Hemost 25,173–82.

104. Markus, G. (1984) The role of hemostasisand fibrinolysis in the metastatic spread ofcancer Semin Thromb Hemost 10, 61–70.

105. Ruf, W., and Mueller, B.M. (1996) Tissuefactor in cancer angiogenesis and metastasisCurr Opin Hematol 3, 379–84.

106. Paredes, N., Xu, L., Berry, L.R., and Chan,A.K. (2003) The effects of chemotherapeutic

132 Mousa

agents on the regulation of thrombin on cellsurfaces Br J Haematol 120, 315–24.

107. Wojtukiewicz, M.Z., Zacharski, L.R.,Memoli, V.A., Kisiel, W., Kudryk, B.J.,Rousseau, S.M., and Stump, D.C. (1990)Fibrinogen-fibrin transformation in situ inrenal cell carcinoma Anticancer Res 10,579–82.

108. Zacharski, L.R., Wojtukiewicz, M.Z.,Costantini, V., Ornstein, D.L., andMemoli, V.A. (1992) Pathways of coagu-lation/fibrinolysis activation in malignancySemin Thromb Hemost 18, 104–16.

109. Hull, R.D., Raskob, G.E., Pineo, G.F.,Green, D., Trowbridge, A.A., Elliott, C.G.,Lerner, R.G., Hall, J., Sparling, T., andBrettell, H.R., et al. (1992) Subcuta-neous low-molecular-weight heparin com-pared with continuous intravenous heparinin the treatment of proximal-vein thrombo-sis N Engl J Med 326, 975–82.

110. Pinedo, H.M., Verheul, H.M., D’Amato,R.J., and Folkman, J. (1998) Involvementof platelets in tumour angiogenesis? Lancet352, 1775–7.

111. Chambers, A.F., MacDonald, I.C., Schmidt,E.E., Koop, S., Morris, V.L., Khokha, R.,and Groom, A.C. (1995) Steps in tumormetastasis: new concepts from intravitalvideomicroscopy Cancer Metastasis Rev 14,279–301.

112. Rickles, F.R., and Levine, M.N. (2001) Epi-demiology of thrombosis in cancer ActaHaematol 106, 6–12.

113. Mousa, S., Inventor Vascular VisionPharmaceuticals, Assignee. Prevention andtreatment of tumor growth, metastasis, andthromboembolytic complications in cancerpatients. United States patent 6,908,907;2005.

114. Levine, M., Hirsh, J., Gent, M., Arnold,A., Warr, D., Falanga, A., Samosh, M.,Bramwell, V., Pritchard, K.I., and Stewart,D., et al. (1994) Double-blind randomisedtrial of a very-low-dose warfarin for preven-tion of thromboembolism in stage IV breastcancer Lancet 343, 886–9.

115. Mousa, S.A., and Mousa, A.S. (2004) Angio-genesis inhibitors: current & future direc-tions Curr Pharm Des 10, 1–9.

116. Folkman, J., and Shing, Y. (1992) Controlof angiogenesis by heparin and other sul-

fated polysaccharides Adv Exp Med Biol 313,355–64.

117. Mousa, S.A., and Mohamed, S. (2004) Anti-angiogenic mechanisms and efficacy of thelow molecular weight heparin, tinzaparin:anti-cancer efficacy Oncol Rep 12, 683–8.

118. Ludwig, R.J., Boehme, B., Podda, M.,Henschler, R., Jager, E., Tandi, C.,Boehncke, W.H., Zollner, T.M., Kaufmann,R., and Gille, J. (2004) EndothelialP-selectin as a target of heparin action inexperimental melanoma lung metastasis Can-cer Res 64, 2743–50.

119. Zhang, Y., Deng, Y., Luther, T., Muller, M.,Ziegler, R., Waldherr, R., Stern, D.M., andNawroth, P.P. (1994) Tissue factor controlsthe balance of angiogenic and antiangiogenicproperties of tumor cells in mice J Clin Invest94, 1320–7.

120. Engelberg, H. (1996) Actions of heparin inthe atherosclerotic process Pharmacol Rev48, 327–52.

121. Borsig, L., Wong, R., Feramisco, J., Nadeau,D.R., Varki, N.M., and Varki, A. (2001)Heparin and cancer revisited: mechanisticconnections involving platelets, P-selectin,carcinoma mucins, and tumor metastasis ProcNatl Acad Sci USA 98, 3352–7.

122. Baughman, R.A., Kapoor, S.C., Agarwal,R.K., Kisicki, J., Catella-Lawson, F., andFitzGerald, G.A. (1998) Oral delivery ofanticoagulant doses of heparin. A ran-domized, double-blind, controlled study inhumans Circulation 98, 1610–5.

123. Hull, R., Kakkar, A., Marder, V., Pineo,G., Goldberg, M., and Raskob, G. (2001)PROTECT trial: oral SNAC-heparin vsenoxaparin for preventing venous throm-boembolism following total hip replacementBlood 100, 148a–9a.

124. Kim, S.K., Lee, D.Y., Lee, E., Lee, Y.K.,Kim, C.Y., Moon, H.T., and Byun, Y. (2007)Absorption study of deoxycholic acid-heparinconjugate as a new form of oral anti-coagulant J Control Release 120, 4–10.

125. Mousa, S.A., Zhang, F., Aljada, A.,Chaturvedi, S., Takieddin, M., Zhang,H., Chi, L., Castelli, M.C., Friedman,K., Goldberg, M.M., and Linhardt, R.J.(2007) Pharmacokinetics and pharmacody-namics of oral heparin solid dosage form inhealthy human subjects J Clin Pharmacol 47,1508–20.

Chapter 4

Laboratory Methods and Management of Patientswith Heparin-Induced Thrombocytopenia

Margaret Prechel, Walter P. Jeske, and Jeanine M. Walenga

Abstract

The clinical effects of heparin are meritorious and heparin remains the anticoagulant of choice for mostclinical needs. However, as with any drug, adverse effects exist. Heparin-induced thrombocytopenia(HIT) is an important adverse effect of heparin associated with amputation and death due to thrombosis.Although the diagnosis and treatment of HIT can be difficult and complex, it is critical that patients withHIT be identified as soon as possible to initiate early treatment to avoid thrombosis.

Key words: Heparin-induced thrombocytopenia, HIT antibodies, detection, diagnosis, thrombininhibitors.

1. PathologicMechanism of HIT

1.1. AntibodyGeneration

HIT is an immune-mediated adverse response to heparintreatment (1–3). HIT antibodies are not directed toward hep-arin specifically, but rather to platelet factor 4 (PF4) in complexwith heparin. PF4 is a positively charged protein stored in thealpha granules of platelets. Exposed lysine and arginine residueson the tetrameric PF4 molecule bind to negatively charged hep-arin molecules (4, 5). This binding exposes cryptic regions withinthe PF4 molecule creating antigenic neoepitopes (6–8). Multi-ple PF4 tetramers arrayed in a lattice with several molecules ofheparin are highly immunogenic and play a fundamental rolein antibody formation (9). Antibodies formed in response tothe heparin:PF4 complex (H:PF4) subsequently recognize PF4bound to cell membranes (10) or other surfaces (11). Antibodies


133

134 Prechel, Jeske, and Walenga

to other heparin-binding proteins, such as neutrophil activatingprotein (NAP2) and interleukin (IL8), have been identified; how-ever, those to PF4 are found in most patients with HIT (12, 13).

Due to its smaller molecular size, low-molecular weight hep-arin (LMWH) has less ability to bind to the PF4 tetramer, alterits configuration, and cause the generation of HIT antibodies.Patients treated with LMWH are 2–3 times less likely to developHIT antibodies than patients treated with unfractionated hep-arin (UFH). However, in vitro studies demonstrate that LMWHcross reacts with existing HIT antibodies formed in response toUFH (14).

1.2. Role of HITAntibodies

HIT antibodies, once formed, become involved in various hemo-static activation processes. Immune complexes of HIT IgG andH:PF4 crosslink platelet FcγIIa receptors, resulting in plateletactivation and release of additional PF4. In the presence ofheparin there is continued formation of antigenic complexes, ini-tiating a cycle of platelet activation and aggregation, and genera-tion of highly procoagulant platelet microparticles (1, 6, 15, 16).Sustained platelet activation contributes to platelet clearance andthrombin generation that can lead to both thrombocytopenia andHIT-associated thrombosis.

Platelets activated by HIT antibodies induce an inflamma-tory state in which macrophages, monocytes, and neutrophils areactivated (15, 17–19). Antibody and leukocyte binding to acti-vated endothelial cells cause the release of tissue factor, plasmino-gen activator inhibitor (PAI)-1, and cytokines, as well as an up-regulation of adhesion molecule expression promoting localizedplatelet and monocyte binding (17, 20–23). Heparan sulfate onthe endothelial cell surface can bind PF4, forming a complex thatis recognized by HIT antibodies (10, 23). The inter-relationshipsof platelets, leukocytes, the endothelium, and the inflammatorystate determine the clinical expression of HIT.

1.3. Seroconversion H:PF4 antibodies are necessary but not sufficient to cause theclinical symptoms of HIT (thrombocytopenia and thrombosis),as many patients who develop HIT antibodies remain asymp-tomatic. HIT can develop from any heparin exposure, includ-ing incidental amounts from heparin flushes or heparin-coateddevices. Frequency of seroconversion and development of throm-bocytopenia and/or thrombosis associated with HIT are variableand depend on factors such as patient population and presence ofcomorbid complications (24–26).

Development of thrombocytopenia with or withoutthrombosis is not always proportional to the incidence ofseroconversion. While 25–50% of cardiac surgery patients formantibodies, less than 2.0% develop the clinical symptoms of HIT.Among orthopedic patients, 15% can be antibody positive but

Lab Methods and Management of HIT Patients 135

only 5% develop clinical consequences (27). Overall, clinicallysymptomatic HIT develops in 1.0% of hospital patients receivingheparin in any form (28).

1.4. AdditionalImmunogenicPotential

The immunogenic potential of heparins is more complex thanthat of protein-derived drugs such as hirudin, aprotinin, FVIIIconcentrate, and erythropoietin. Because of their polycomponentnature and their interactions with multiple endogenous proteins(e.g., fibronectin, growth factors, serpins), heparins and relatedpolysaccharides are likely to generate an array of antibodies. Someof these antibodies may modulate the pharmacologic actionsof heparin. Thus, apart from the so-called HIT antibodies andtheir associated thrombosis, patients generating heparin-mediatedantibodies may exhibit a therapeutic compromise requiringdose adjustment of the heparin or an alternative approach foranticoagulation.

2. Clinical HIT

2.1.Thrombocytopenia

HIT is typically described as an otherwise unexplained throm-bocytopenia starting 4–14 days after administration of hep-arin. Thrombocytopenia is usually defined as a platelet count<100–150,000×109/L; however, HIT may also be recognizedby a 30–50% drop from the pre-heparin baseline even if theplatelet count remains above this threshold. No single defini-tion of thrombocytopenia is appropriate in all clinical situations(29). HIT is particularly difficult to diagnose in patient popula-tions where low platelet counts are typical. In orthopedic and car-diac surgery patients, HIT may be recognized by the pattern ofplatelet count recovery or by a particular percent decrease com-pared to the post-surgical platelet level. Patients requiring ven-tricular assist devices who receive anticoagulation during surgeryand for extended post-operative periods of mechanical circulatorysupport often develop H:PF4 antibodies, but they also have mul-tiple explanations for low platelet counts (30). No guidelines areyet established for HIT in this patient group.

In a patient with previous heparin exposure, particularlywithin the past 120 days, H:PF4 antibodies may already bepresent and lead to early onset of HIT, even within hours of thenext heparin exposure. On the other hand, patients may expe-rience a late drop in platelet count. If the patient is dischargedfrom the hospital within several days after exposure to heparin,the thrombocytopenia will go undetected, and the diagnosis ofHIT-associated thrombosis could potentially be missed.


2.2. Thrombosis Thrombosis occurs in approximately 35% of HIT patients withthrombocytopenia. Thrombosis can be apparent at the time ofHIT diagnosis, others can develop thrombosis within the next30 days if not treated with a non-heparin anticoagulant. HIT isassociated with a wide spectrum of arterial and venous throm-boembolic complications (e.g., deep vein thrombosis, pulmonaryembolism, myocardial infarction, thrombotic stroke, ischemiclimb, vein graft occlusion, skin lesions at injection sites). Mortal-ity among patients with HIT thrombosis is as high as 30%, with20% of those surviving requiring a limb amputation.

2.3. Scoring Systems Diagnosis of HIT is based primarily on clinical presentation.Algorithms and diagnostic scoring systems have been proposedto help clinicians evaluate clinical impressions based on the tim-ing and extent of thrombocytopenia and the presence or absenceof thrombotic complications or other explanations for low plateletcount (31–33). The performance of such risk assessment strate-gies varies with the experience of the clinician (34). A separatescoring system has been proposed for use in patients followingcoronary artery bypass surgery (35). This score takes into accountthe platelet count time course, the time between surgery and theindex date and the duration of surgery. In a retrospective study,changes in the clinical score between initial testing and subse-quent testing in patients with an initial negative ELISA were pre-dictive of the development of HIT with thrombosis (36).

Such scoring systems offer the possibility of focusing lab test-ing on those patients most likely to develop clinical HIT andto identify those most likely to benefit from alternate anticoag-ulant therapy. Further prospective clinical studies are needed todetermine the usefulness of such scoring systems in guiding HITmanagement.

3. LaboratoryMethods

There are two types of laboratory tests for HIT (Table 4.1).Antigen assays detect the presence of immunoglobulins thatbind the antigenic neoepitopes exposed in H:PF4 complexes.Functional assays demonstrate platelet activation caused by HITantibody immune complexes. Each type of test provides uniqueinformation and should not be used as the sole basis to diag-nose HIT.

3.1. Antigen Assays Antigen assays for HIT detect antibodies that recognize andbind cryptic PF4 epitopes. Two solid phase enzyme-linked


Table 4.1Laboratory tests for the diagnosis of HIT

Platelet count monitoring

• The frequency of monitoring depends on the patient’s risk fordeveloping HIT

• General recommendations:

– Monitor platelet count daily for UFH– Monitor platelet count every other day for LMWH

ELISA-type assays detect the presence of antibodies to the H:PF4 complex• Highly sensitive but not specific

• May not be predictive of clinical HIT (“false positive”)• May not correlate with a positive response in the platelet function

assay• Long-term clinical implications of antibody positivity unclear (i.e.,

patients who do not develop HIT clinical symptoms except forantibody generation)

Platelet function assays for HIT antibodies

• Good association with HIT thrombocytopenia and thrombosis• Two primary types of assays:

– 14C-serotonin release assay (SRA; uses washed platelets)– Heparin-induced platelet aggregation test (PAT; uses platelet-rich

plasma)

Each type of laboratory assay for HIT provides unique informationEach type of laboratory assay for HIT differs in sensitivity and specificity

Knowledgeable use and interpretation of each test are importantMultiple testing over several days improves chance of identifying a HIT

patient

Combined results from ELISA, SRA, and PAT improves chance ofidentifying a HIT patient

Laboratory test results should not be used to guide initial therapeuticdecisions, but rather to confirm a clinical diagnosis of HIT to guidefuture therapy

immunosorbent assays (ELISAs) are commercially available. Thecapture probe of the Asserachrom ELISA (Stago) is heparinin complex with recombinant human PF4; the GTI HATELISA (Genetics Testing Institute) utilizes negatively chargedpolyvinylsulfonate bound with PF4 from human platelets. Resultsare expressed as optical density (OD); OD values for pos-itive/negative thresholds are provided by the manufacturers.Because HIT antibodies are polyclonal and they differ in speci-ficity and affinity, it is not surprising that these two ELISA kitsdetect slightly different cohorts of H:PF4 antibodies (37), whereopposite results are obtained in 15% of patient samples (38, 39).


These ELISA assays utilize a global conjugate reagent capableof detecting IgG, IgA, and IgM type HIT antibodies. ELISAsare now being offered that detect only HIT IgG. This is basedon studies which demonstrated that only IgG isotype HIT anti-bodies activate platelets (40, 41). While the IgG-specific assay ismore sensitive for overt HIT where patients have marked throm-bocytopenia and thrombosis, the global assay may provide earlierevidence of a progressing immune reaction.

It is recommended to also test patient specimens in the pres-ence of high concentration heparin for a more accurate diagno-sis of HIT. Addition of high heparin to the ELISA test reactioninhibits HIT antibody binding by removing or blocking accessto antigenic PF4 epitopes (2). Thus, true HIT antibodies willnot bind in the presence of high heparin, and thus there will be anegative response with high heparin. Antibodies with non-specificbinding will have positive OD values in the presence as well as inthe absence of high heparin. There is data to indicate that whenantibody binding was not inhibited by high heparin, patients wereunlikely to have HIT, suggesting that the extra step provides use-ful diagnostic information (42).

Two antigen-based rapid detection HIT antibody assays arealso available. The particle gel immunoassay (PaGIA) is an agglu-tination assay which utilizes polystyrene beads coated with H:PF4complexes and test serum/plasma incubated in a chamber of anID-MicroTyping test card (DiaMed, Switzerland). Centrifugationof the test card separates the beads cross-linked by H:PF4complexes which are identified by visual inspection. A morequalitative assessment can be done by testing serial dilutions ofspecimen and reporting antibody titer in terms of the highest dilu-tion showing a positive result (43). The particle immunofiltrationassay (PIFA) (Akers Biosciences, Thorofare, NJ) uses microparti-cles coated with PF4 within a self-contained minireactor device.Addition of (non-frozen/thawed) serum containing H:PF4 anti-bodies will cause matrix formation and trap the microparticleswithin the chamber membrane. Non-matrixed microparticles, inHIT antibody negative specimens, migrate through the mem-brane and are detected in the test result window of the device.Clinical experience with these point-of-care devices has beenmixed (44–46).

3.2. FunctionalAssays

The functional tests are bioassays that utilize fresh plateletsfrom a known reactive normal donor. Addition of heparin toplatelets/patient serum (PF4 present) allows H:PF4 complexes toform and present the HIT antigen. HIT antibodies in the patientserum bind to the H:PF4 complex and cause platelet activation.

The serotonin release assay (SRA) is conducted with plateletsthat have been incubated with 14C-radiolabeled serotonin, thenwashed and resuspended in calcium-containing buffer. Plateletsare Fcγ receptor-bearing cells that are activated by IgG immune


complexes. Platelet activation resulting in granule release isdetected by the presence of radioactivity in the incubationsupernatant. By measuring background radioactivity and total14C-serotonin uptake, the strength of the activation responseto HIT IgG is quantified as percent serotonin release. Plateletactivation is usually defined as 20% or greater serotonin release.Stronger release activity presumably indicates high affinity or hightiter antibody; 50–80% release has been shown to be more spe-cific for HIT patients with thrombocytopenia and/or thrombo-sis (31).

The other commonly used functional test for the clinicallaboratory diagnosis of HIT is the platelet aggregation test(PAT) for HIT antibodies. This test utilizes donor platelets(prepared as platelet rich plasma), heparin, and patient serumand is performed on a commercial aggregometer where plateletaggregation is measured by an increase in light transmission.Tests conducted in platelet-rich plasma are considered less sen-sitive than washed platelet assays; however, patient specimenscan test positive by PAT but negative by SRA and vice versa(32, 47).

For both the SRA and the PAT, an important control toincrease the assay specificity is inclusion of both low (0.1 U/ml)and high (100 U/ml) heparin concentrations. A positive resultin a HIT diagnostic assay is platelet activation in the presence oflow heparin, but not in the presence of high heparin. An acti-vation assay result is “indeterminate” when a specimen causesplatelet activation at both low and high concentrations of hep-arin, which indicates that the antigenic target is not heparin-dependent. These specimens may contain pre-formed immunecomplexes or antibodies such as anti-HLA or anti-platelet gly-coprotein antibodies.

There is considerable donor-related variability in plateletresponsiveness in these functional platelet assays, making it imper-ative that tests are performed with platelets from known reactivedonors (48, 49). Each assay should also include known HIT anti-body positive and negative control sera. The functional tests forHIT are highly complex, difficult to standardize, and require care-ful attention to quality control measures. The most reproducibleresults are obtained when these assays are conducted in experi-enced reference laboratories (50).

Other platelet function assays include the heparin-inducedplatelet aggregation (HIPA) assay which utilizes washed plateletsand a visual assessment of platelet aggregation over time in amicrotiter plate (51). The strength of the activation response inthis assay is reflected in the lag time until aggregation is observed.HIT antibody immune complexes can also be detected by ADPrelease measured by lumi-aggregometry or by flow cytometrywhere either platelet microparticle formation or annexin bindingis detected (16, 52–54).


3.3. AssayResponses

3.3.1. ELISA Specificity Many specimens that test positive by ELISAs do not cause plateletactivation in the SRA or PAT and are not associated with throm-bocytopenia or thrombosis. Thus, the ELISA tests are not veryspecific for HIT (40).

On the other hand, a negative ELISA result can rule-out HITin patients with a low clinical probability of HIT based on clinicalsymptoms (55). For patients with a negative ELISA test result butwho are clinically suspected of having HIT, it is useful to performa combination of different types of tests for HIT antibodies and torepeat testing over a period of several days to assure that antibodydoes not develop.

3.3.2. HIT PositivePatients

The most straightforward interpretation of HIT tests is on symp-tomatic patients with thrombocytopenia and/or thrombosis dur-ing the period 5–14 days post-heparin exposure. When assessmentof the platelet count and/or thrombotic symptoms give rise toreasonable suspicion of HIT, a positive lab result can affirm thediagnosis of HIT and justify initiating or maintaining alternativeanticoagulant therapy. When clinical suspicion is high, one neg-ative test should not rule-out HIT. Repeat testing or use of anadditional type of test is advisable, along with careful surveillanceof platelet counts.

The interpretation and value of HIT diagnostic tests vary withthe timing of the collection of the patient specimen and the clin-ical status of the patient. Pre-operative testing may be useful insurgical patients with recent heparin exposure or previous historyof clinically symptomatic HIT or in patients with inflammatoryor malignant complications. Assuming a negative pre-operativeantibody result, tests within 1–4 days after heparin exposure pro-vide little information, since H:PF4 antibodies would not yet beapparent.

In the clinical setting, both platelet activation and anti-gen tests are typically reported as either positive or negative.Retrospective studies suggest that the magnitude of a positiveresponse might be helpful in making a diagnosis of HIT. Patientswith stronger activation results (50–80% serotonin release)and/or higher antibody titers (OD 1.0–1.2) have a greater likeli-hood of having clinically symptomatic HIT (56–58). When eval-uating weaker assay responses, it is important to remember thatresults may be different on a repeat, subsequent test, particularlyin those patients whose OD value was in the upper portion of thenegative range (35, 59, 60). While it is advisable to follow thisprocess for results from the ELISA tests, results from the plateletfunction tests can be more difficult to interpret day to day due tothe variabilities observed with platelet activity.


3.3.3. Post-operativePatients

Interpretation of test results, particularly in ELISA assays, inpatients without clinical symptoms of HIT during the post-heparin interval is problematic. Many patients have a positiveresponse but do not develop thrombocytopenia or thrombosis.Other patients may have a negative response, yet develop symp-toms and a positive test result on subsequent days (e.g., delayed-onset HIT). Thus, in this population, the ELISA test result needsto be interpreted with caution.

3.3.4. Additional RisksAssociated withAntibodies

There is a growing body of evidence to suggest that even in theabsence of thrombocytopenia, anti-H:PF4 antibody generation isassociated with an increased risk of thrombotic events (61). Thisis most clearly seen in cardiovascular patients who have receivedheparin for the treatment of acute coronary syndrome or duringcoronary bypass surgery. In these studies, a significant increasein the incidence of typical endpoints such as death, myocardialinfarction, recurrent angina, the need for urgent revasculariza-tion, and post-operative length of stay was observed in patientswho were antibody positive compared to those who were neg-ative (24, 62–64). This increased risk is more subtle in non-cardiac patients such as those undergoing hemodialysis, vascularsurgery, or post-surgical deep vein thrombosis (DVT) prophy-laxis (65–68). Until there is more information on risks associ-ated with HIT-seropositivity itself, however, routine screening ofpatients without thrombocytopenia or unexplained thrombosis isnot recommended.

4. ClinicalManagement

The management of patients with HIT consists of high clinicalawareness, early diagnosis, and early treatment. Once HIT is sus-pected, there is a necessity for immediate intervention to initi-ate anticoagulation treatment against the high risk of thrombosis(Table 4.2). Cessation of heparin alone is not sufficient to removethe threat of thrombosis (69). One should not wait for laboratoryresults to act. Also initial therapeutic decisions should not dependupon a positive laboratory test, but should be based upon clini-cal findings (i.e., thrombocytopenia and/or new thromboembolicevents). It is important, however, to have laboratory confirmationof HIT since patients who have HIT are at great risk for recur-rence should they be exposed to heparin in the future (70). Inother words, lab tests should not be used to guide initial thera-peutic decisions but rather to confirm a clinical diagnosis of HITand guide future therapy.


Table 4.2Anticoagulant treatments for patients with HIT

Direct thrombin inhibitors (DTI)

• Argatroban and lepirudin are the treatments of choice for management of HIT thrombosis• Bivalirudin is under development

• Monitor with aPTT• Each DTI has individual pharmacologic characteristics

Danaparoid (FXa inhibitor)• Successfully used for over 10 years for the management of HIT

• Monitor platelet counts at onset of administration to check for clinically significant cross-reactivityto HIT antibodies

Vitamin K Antagonists

• Used for long-term thrombosis management• Follow guidelines to facilitate cross-over between the DTI and the VKA to avoid coumadin-induced

necrosis and monitoring issues

For interventional cardiology procedures in patients with HIT• Argatroban is approved

• Bivalirudin is approved for use in non-HIT patients• Monitor with ACT

For cardiac surgery in patients with HIT• Best to use UFH; best that it is more than 3 months since patient last exposed to heparin; test to

assure that patient is HIT antibody negative

• For patients with HIT antibody titer requiring life-saving surgery, alternative anticoagulant can beused but dosing and monitoring regimens are not optimized

Potential other anticoagulant choices (future)

• Fondaparinux (FXa inhibitor) is not approved for HIT but has been successfully used in some HITpatients; some issues with antibody formation and cross-reactivity have occurred

• Idraparinux

• Oral FXa inhibitors• Oral DTIs

• Other agents• Drug combinations to provide multiple targeted antithrombotic protection

LMWH is contraindicated in patients with HIT because it hasa high rate of interacting with established HIT antibodies. Platelettransfusions are also contraindicated in patients with HIT.

4.1. Direct ThrombinInhibitors (DTIs)

DTIs are strong anticoagulants that inhibit the high level ofthrombin generation in patients with HIT. Because the chemicalstructures of the DTIs are different from that of heparin, thesedrugs do not generate HIT antibodies nor do they interact withpre-formed HIT antibodies.


Treatment with a DTI significantly reduces the risk ofthrombosis and thromboembolic complications (new thrombosis,amputation, death) associated with HIT (71–74). Significantlymore treated patients remained event free and platelet countsrecovered more rapidly in patients receiving DTI treatment.

Due to the inherent bleeding risk with all DTIs, it is impor-tant to monitor treatment. Particular attention should be given toelderly patients and patients with renal/liver failure. The activatedpartial thromboplastin time test (aPTT) has been recommendedas the monitoring assay. In general DTIs are dosed to a 2–2.5-foldincrease in the aPTT.

Laboratory tests that use a clotting endpoint, such as fib-rinogen and coagulation factor assays, will be affected by a DTI“contaminant” in the patient’s specimen (75, 76). True factorlevels can be measured by chromogenic or immunologic basedassays which are not affected by the DTI.

Although there are many similarities among the DTIs, impor-tant differences exist between them (Table 4.3) (70, 77, 78). Thedifferent chemical structure of each DTI defines where each drugbinds to thrombin, the tightness of the binding, and so on. Thesecharacteristics are reflected in the different pharmacokinetic andpharmacodynamic behavior of each drug.

4.1.1. Argatroban Argatroban has been approved by the health authorities of theUnited States, Canada, and Europe for both the prophylaxis andthe treatment of HIT thrombosis. It is hepatically metabolizedand thus the anticoagulant of choice in patients with renal fail-ure (79). Argatroban is administered by continuous infusion ata dose of 1.7–2.0 μg/kg/min for 5–7 days (71, 72). Plasmaargatroban levels rapidly decline in about 40 min when drug isdiscontinued, and coagulation parameters generally return to pre-treatment values within 2–4 h (80–82). Repeated exposure toargatroban does not generate antibodies to argatroban, bleeding,or other adverse events (83, 84). Argatroban can be distinguishedfrom other DTIs in that it produces an increase in nitric oxide,which may contribute to its therapeutic efficacy by modulatingvascular and cellular function (85).

Argatroban has also been approved by the US Food and DrugAdministration (FDA) for anticoagulation of HIT patients dur-ing percutaneous coronary interventions (PCI) (86). There arereports on the successful use of argatroban anticoagulation inpediatric patients requiring interventional cardiology proceduresor extracorporeal life support (87, 88) and for stent implanta-tion in renal arteries (80, 89, 90). Argatroban anticoagulation inPCI (at a reduced dose) used in combination with glycoproteinIIb/IIIa inhibition for PCI is well tolerated with an acceptablebleeding risk (91).


Table 4.3Comparison of the pharmacologic responses of the commonly used alternativeanticoagulants for the management of HIT thrombosis

Argatroban (DTI)

• Approved for both prophylaxis and treatment of HIT thrombosis• Approved for use in interventional cardiology procedures in HIT and non-HIT patients

• Anticoagulant response rapidly reversed (40–50 min)• Cleared through the liver

• Adjust dose in hepatically impaired patients• Affects the INR

• Antibodies are not generated against argatrobanLepirudin (DTI; hirudin)

• Approved only for treatment of HIT thrombosis• Anticoagulant response slowly reversed (1.3–3 h)

• Bleeding tends to be higher with lepirudin than with other DTIs• Cleared through the kidneys

• Adjust dose in renally impaired patients• Antibodies are generated in 45% of treated patients; upon re-exposure severe anaphylaxis and death

have occurred

Danaparoid (FXa inhibitor)• Has been widely used in North America and Europe

• Long elimination half-life (25 h)• Can be given either intravenously or subcutaneously

• Low bleeding risk• Mainly cleared through the kidneys

• Routine monitoring not required; monitoring required in patients with excessively low or highbody weight or renal failure (use anti-FXa assay)

• Small potential for clinically relevant cross-reactivity with HIT antibodies; platelet counts shouldbe monitored at initiation

• Multiple mechanisms of antithrombotic action, including an anti-inflammatory effect

In a more recent study, argatroban was found to effectivelyreduce new stroke and stroke-associated mortality in patients withHIT without increasing intracranial hemorrhage (92).

4.1.2. Lepirudin Lepirudin has been approved for the treatment of HIT throm-bosis by the health authorities of the United States, Canada, andEurope. It is an irreversible inhibitor of thrombin with an elim-ination half-life of about 90 min (1.3–3 h). Lepirudin is renallyexcreted and needs to be used with caution in patients with renalimpairment. Intravenous dosing ranges from 0.1–0.4 mg/kg/h,with or without an initial bolus, for 11–14 days in HIT patients


(73, 74). Bleeding rates tend to be higher with lepirudin thanwith other DTIs.

Exposure to lepirudin results in antibody formation in abouthalf of the treated patients (93–95). These antibodies alter thepharmacokinetics of lepirudin necessitating careful monitoring toavoid bleeding complications. Re-exposure to lepirudin has beenlinked to at least nine reported cases of severe anaphylaxis with atleast five fatal outcomes (96).

Lepirudin has been studied in HIT patients undergoing PCI(97), but it has not been approved for this setting.

4.1.3. Bivalirudin Bivalirudin is a reversible inhibitor of thrombin that is largelyrenally excreted. Although not approved by the FDA, bivalirudinhas been used to anticoagulate patients with HIT thrombosis.Bivalirudin is, however, approved for use in PCI in non-HITpatients (98).

Perhaps the greatest obstacle to overcome in the managementof patients with HIT is anticoagulation during surgical coronaryrevascularization or heart valve replacement surgery. For patientswith active HIT, anticoagulation with bivalirudin was shown tobe feasible in both on-pump (cardiopulmonary bypass, CPB) andoff-pump (OPCAB) cardiac surgery (99, 100). However, the useof any DTI in cardiac surgery is associated with inherent risksbecause there is no antidote for any DTI. Dosing guidelines havenot been fully established and bleeding can be excessive, andmonitoring the high drug levels is an unresolved issue. Completeefficacy against blood clotting is also a concern.

4.1.4. Oral ThrombinInhibitors

The oral DTI dabigatran was approved March 2008 for theprevention of venous thromboembolic events in patients whohave undergone total hip replacement surgery or total kneereplacement surgery in Europe. This drug can be consideredfor anticoagulation of patients with HIT thrombosis. Althoughthis small molecule, direct acting, thrombin inhibitor has notbeen studied in HIT, based on its chemical structure, it is notexpected to bind PF4 or have any interaction with pre-formedHIT antibodies.

4.2. Factor XaInhibitors

4.2.1. Danaparoid Danaparoid has been used to successfully treat HIT patientsfor more than 10 years (101–103). Because danaparoid has alow bleeding risk, routine monitoring is not required, except inpatients with excessively low or high body weight or renal failure(104). It has a sustained effect and can be given either intra-venously or subcutaneously. There is a small potential for clini-cally relevant cross-reactivity of danaparoid with HIT antibodies


in patients, so platelet counts should be monitored during theinitial phase of treatment (105, 106).

A potential advantage of danaparoid over the DTIs isthat it has multiple mechanisms of action, including an anti-inflammatory effect. Thus in addition to inhibition of the coag-ulation system, danaparoid may be able to affect other aspects ofthe pathophysiology of HIT.

4.2.2. Fondaparinux The synthetic derivative of heparin, fondaparinux, is of very lowmolecular weight and does not bind to PF4. Thus, it may be a use-ful alternative anticoagulant for HIT patients. In fact, there arereports of its successful use in several HIT patients (107–110).However, there is also one report of HIT-induced thrombo-sis associated with fondaparinux treatment (111) and anotherreport of fondaparinux interacting with LMWH-induced HITantibodies (112). Furthermore, there is evidence that HIT-typeantibodies are generated with fondaparinux treatment (113). Atthis time and until more is known, fondaparinux should be usedwith caution in patients with HIT.

4.2.3. Idraparinux Idraparinux is a sister molecule of fondaparinux. Its extended half-life allows for once per week dosing. This agent is in clinical trialfor prophylaxis against venous thrombosis. As fondaparinux hasshowed success in a limited number of patients with HIT, idra-parinux may also be considered for use in HIT; however, noreports have been published. Other structurally modified deriva-tives of fondaparinux, such as non-PF4 binding agents, which arein development, may be of interest in the future.

4.2.4. Oral XaI Inhibitors Small molecule, direct acting, factor Xa inhibitors that canbe orally administered include rivaroxaban and apixaban. Bothhave been studied in clinical trials for prophylaxis againstvenous thrombosis following orthopedic surgery (114–117).In September 2008, rivaroxaban was approved in Europe andCanada, and continues to seek approval from the US FDA.Apixaban is awaiting approval in several countries. These agentscan be considered for use in the management of thrombosis inpatients with HIT because they do not bind PF4, and they shouldhave no interaction with pre-formed HIT antibodies (118). Theyalso should not generate HIT antibodies. Although no clinicalinformation is yet available, if these oral agents are found to beuseful for the clinical management of HIT, an advantage wouldbe their application for both acute and long-term treatment.

4.3. Vitamin KAntagonists

For long-term anticoagulation of HIT patients, vitamin K antag-onists (VKAs) are used after the treatment period with a DTI.Specific dosing guidelines for the VKA need to be fol-lowed to avoid thrombotic complications in patients with HIT(29, 70, 119). VKA can be initiated when the patient is outof the acute phase of HIT (i.e., platelet count on the rise and


>100×109/L). It should be started at a low dose (a loading doseshould not be used) while the patient is fully anticoagulated witha DTI. The DTI can be tapered off when the INR for the VKAis therapeutic and stable. VKA treatment should continue untilplatelet counts recover to a stable plateau or longer if clinicallywarranted.

DTIs prolong the PT/INR (75, 120–123). INRs >5 com-monly occur with argatroban–warfarin co-therapy but this doesnot correspond to an effect on coagulation factor levels and bleed-ing is not enhanced (77, 122). There is a predictable linear effectof argatroban doses up to 2 μg/kg/min on the INR duringwarfarin co-therapy. This allows for reliable prediction of the levelof oral anticoagulation with the VKA (121). To transition fromlepirudin to VKA, the dose of lepirudin is first reduced to an aPTTratio just above 1.5. Lepirudin is then continued for 4–5 dayswith the VKA and discontinued when the INR is therapeutic(73, 74).

4.4. FurtherConsiderations

4.4.1. Cardiac Surgery Until a DTI or another anticoagulant is approved for use incardiac surgery, heparin remains the safest and most effective anti-coagulant in this high-risk clinical setting. However, subsequentuse of heparin after resolution of HIT can be hazardous particu-larly within the first 3 months. A brief exposure to heparin dur-ing surgery can be considered under compelling circumstancesfor patients with a history of HIT who have HIT antibodies thatare not detectable by a functional platelet assay (70). In this cir-cumstance, standard heparin protocols restricted to the surgeryitself can be employed with the use of a DTI or a VKA for post-operative care (70, 124).

4.4.2. Multi-TargetedTreatment

Although DTIs have made a huge advance in the clinicalmanagement of patients with HIT thrombosis, there remains anunacceptable rate of morbidity and mortality in this patient pop-ulation. If one considers the pathophysiology of HIT, it seemsobvious that inhibition of thrombin, while important, cannotprovide complete antithrombotic management of HIT throm-bosis. HIT is associated not only with a hypercoagulable state,but also platelet activation, vascular endothelial dysfunction, andinflammation (leukocyte activation, cytokine up-regulation) (17,21, 23).

HIT is a multi-pathologic condition that may be best treatedby multi-targeted therapy. Limited studies suggest that a combi-nation of an inhibitor of thrombin/thrombin generation plus ananti-platelet drug such as GPIIb/IIIa inhibitor (125), not aspirin(126), may be of interest. This combined therapy targets theprinciple mechanisms related to the pathology of HIT: platelet


activation and thrombin generation (15, 127). Combinationsof a thrombin inhibitor and an FXa inhibitor have also beensuggested.

4.4.3. Other PotentialTreatment Drugs

Other antithrombotic drugs that can be considered for themanagement of HIT thrombosis include recombinant thrombo-modulin being developed for disseminated intravascular coagula-tion (DIC), defibrotide, a single-stranded deoxyribonucleic acidderivative used to treat venous occlusive disease, and suledoxide,a heparinoid complex under development for diabetic nephropa-thy. Although these drugs have not been studied for efficacy orsafety in patients with HIT, they have the potential to be useful.

5. RegulatoryConsiderations

An interesting development has recently occurred with the intro-duction of generic LMWHs. Regulatory bodies are challengedto develop specific guidelines for generic LMWH approval dueto the complex nature of these polycomponent biologicals. TheUS FDA has identified immunogenicity of LMWHs as an impor-tant criterion to differentiate LMWHs and to demonstrate thebioequivalence of generic LMWHs to the branded products(128). It remains important to determine how best to demon-strate this equivalence. Demonstration of similar in vitro cross-reactivity is likely to be insufficient. Rather, well-designed clinicaltrials demonstrating a similar propensity to trigger anti-H:PF4antibodies would be preferred.

The immunologic profile of both UFH and LMWH islargely dependent on the heparin components and the pres-ence of other glycosaminoglycans (GAGs) such as dermatans.The starting material for the heparins is heterogeneous, anddue to increased demand, the quality issues for heparin arecompromised. Therefore, even with UFH the prevalence of HITantibodies and their composition may vary and should be furtherinvestigated. If the proposed US FDA guidelines along with theproposed PCR method (129) are implemented, such differencescan be minimized. Because of this, monitoring both UFH andLMWH for HIT is important.

Also of concern is the recent issue of heparin contamina-tion by over-sulfated chondroitin sulfate (OSCS) (130, 131). It isknown that other highly sulfated GAGs and GAG-like moleculeselicit a strong cross-reactivity in in vitro functional assays forHIT. Preliminary data suggest that a similar profile is observedwith OSCS. Additionally, the mixture of OSCS and heparin, orthe LMWH enoxaparin mixed with OSCS, exhibits a differentinteraction compared to the heparin alone. Issues related to this


contaminant are yet to be resolved, and concern exists for poten-tial other contaminants in the future.

While published reports have identified OSCS as the primarycontaminant, other impurities in the recalled batches of heparinhave also been found. Some of these relate to GAGs and are yetto be characterized. Moreover, the molecular profile of OSCSisolated from different contaminated heparins also differs. Someof the recalled batches of LMWHs have been found to containOSCS with varying molecular weight, suggesting that the parentcontaminant in the starting material is depolymerized during themanufacturing process. The presence of a low-molecular weightOSCS in LMWH is of concern as this material binds much morestrongly with PF4 and other proteins and could potentially gen-erate additional forms of antibodies whose functions are currentlyunknown. Therefore, proper monitoring of the contaminant andother impurities in LMWHs is critically important.

6. Summaryand Conclusions

HIT is an immune-mediated serious adverse effect of heparinexposure that results in platelet activation, inflammation, andan extreme hypercoagulable state. HIT antibodies cause plateletactivation, platelet aggregation, the generation of procoagulantplatelet microparticles, and activation of leukocytes and endothe-lial cells. The clinical manifestations are thrombocytopenia and ahigh rate of thromboembolic complications that can progress toamputation or death.

Early diagnosis based on a comprehensive interpretation ofclinical and laboratory information improves clinical outcomes.Careful monitoring for thrombocytopenia and thrombosis duringand for at least several days following heparin treatment of anydose and duration is important.

Limitations of the laboratory assays and atypical clinical pre-sentations often make the diagnosis of HIT difficult. The ELISAand platelet function laboratory tests for HIT provide differentinformation. The ELISA tests merely provide evidence of thepresence of HIT antibodies that may or may not be clinically rel-evant. Thus, specificity of ELISA tests is low. Platelet functiontests detect HIT antibodies that cause platelet activation and havea better correlation to patients with HIT-associated thrombocy-topenia and thrombosis. Because of a lower sensitivity, a negativeplatelet function test result cannot exclude HIT. A combinationof tests repeated over several days provides the best information.

All heparins, including LMWH, must be stopped whenthe diagnosis of HIT is suspected. For the management ofHIT thrombosis, the DTIs argatroban or lepirudin, or the FXainhibitor danaparoid, are recommended. Differences between


drugs need to be considered when making a clinical treatmentdecision for thrombosis prophylaxis, thrombosis treatment, andinterventional procedures. For long-term management of HITthrombosis VKAs are used, but specific dosing guidelines mustbe followed. Bivalirudin and fondaparinux may be other potentialtreatment options but these have not been fully studied in thispatient population. In the future, idraparinux or the oral FXa orthrombin inhibitors may prove to be useful if clinically validated.

The diagnosis and treatment of HIT is complex, but needs tobe considered in the clinical management of patients exposed toheparin due to its serious outcomes. Research and clinical studieswill continue to address the unresolved issues and unmet clinicalneeds associated with HIT.

References

1. Kelton, J.G., Sheridan, D., Santos, A.,Smith, J., Steeves, K., Smith, C., Brown, C.,and Murphy, W.G. (1988) Heparin-inducedthrombocytopenia: laboratory studies Blood72, 925–30.

2. Amiral, J., Bridey, F., Dreyfus, M.,Vissac, M., Fressinaud, E., Wolf, M., andMeyer, D. (1992) Platelet factor 4 complexedto heparin is the target for antibodies gener-ated in heparin-induced thrombocytopeniaThromb Haemost 68, 95–6.

3. Greinacher, A., Potzsch, B., Amiral, J.,Dummel, V., Eichner, A., and Mueller-Eckhardt, C. (1994) Heparin-associatedthrombocytopenia: isolation of the antibodyand characterization of a multimolecularPF4-heparin complex as the major antigenThromb Haemost 71, 247–51.

4. Maccarana, M., and Lindahl, U. (1993)Model of interaction between platelet factor4 and heparin Glycobiology 3, 271–7.

5. Mayo, K.H., Ilyina, E., Roongta, V.,Dundas, M., Joseph, J., Lai, C.K.,Maione, T., and Daly, T.J. (1995) Heparinbinding to platelet factor-4. An NMR andsite-directed .mutagenesis study: arginineresidues are crucial for binding Biochem J312, 357–65.

6. Newman, P.M., and Chong, B.H. (2000)Heparin-induced thrombocytopenia: newevidence for the dynamic binding of puri-fied anti-PF4-heparin antibodies to plateletsand the resultant platelet activation Blood 96,182–7.

7. Suh, J., Aster, R.H., and Visentin, G.P.(1998) Antibodies from patients withheparin-induced thrombocytopenia/throm-bosis recognize different epitopes on hep-arin:platelet factor 4 Blood 91, 916–22.

8. Visentin, G.P. (1999) Heparin-inducedthrombocytopenia: molecular pathogenesisThromb Haemost 82, 448–56.

9. Rauova, L., Poncz, M., McKenzie, S.E.,Reilly, M.P., Arepally, G., Weisel, J.W.,Nagaswami, C., Cines, D.B., and Sachais,B.S. (2005) Ultralarge complexes of PF4and heparin are central to the pathogenesisof heparin-induced thrombocytopenia Blood105, 131–8.

10. Visentin, G.P., Ford, S.E., Scott, P.J.,and Aster, R.H. (1994) Antibodies frompatients with heparin-induced thrombocy-topenia/thrombosis are specific for plateletfactor 4 complexed with heparin or bound toendothelial cells J Clin Invest 93, 81–8.

11. Newman, P.M., and Chong, W.J. (1999)Further characterization of antibody andantigen in heparin-induced thrombocytope-nia Br J Haematol 107, 303–9.

12. Amiral, J., Marfaing-Koka, M., Wolf, M.,Alessi, M.C., Tardy, B., Boyer-Neumann, C.,Vissac, A.M., Fressinaud, E., Poncz, M., andMeyer, D. (1996) Presence of autoantibod-ies to interleukin-8 or neutrophil-activatingpeptide-2 in patients with heparin-associatedthrombocytopenia Blood 88, 410–6.

13. Regnault, V., deMaistre, E., Carteaux, J.,Gruel, Y., and Nguyen, P. (2003) Plateletactivation induced by human antibodies tointerleukin-8 Blood 101, 1419–21.

14. Walenga, J.M., Prechel, M.M., Jeske, W.P.,and Bakhos, M. (2005) Unfractionated hep-arin compared with low-molecular-weightheparin as related to heparin-induced throm-bocytopenia Curr Opin Pulm Med 11,385–91.

15. Jeske, W.P., Walenga, J.M., Szatkowski, E.,Ero, M., Herbert, J.M., Haas, S., and


Bakhos, M. (1997) Effect of glycoproteinIIb/IIIa antagonists on the HIT seruminduced activation of platelets Thromb Res88, 271–81.

16. Warkentin, T.E., Hayward, C.P., Boshkov,L.K., Santos, A.V., Sheppard, J.A., Bode,A.P., and Kelton. J.G. (1994) Sera frompatients with heparin-induced thrombocy-topenia generate platelet-derived micropar-ticles with procoagulant activity: anexplanation for the thrombotic complica-tions of heparin-induced thrombocytopeniaBlood 84, 3691–9.

17. Walenga, J.M., Jeske, W.P., Prechel, M.M.,and Bakhos, M. (2004) Newer insightson the mechanism of heparin-inducedthrombocytopenia Semin Thromb Hemost30(Suppl 1), 57–67.

18. Jeske, W.P., Vasaiwala, S., Schlenker, R.,Wallis, E., and Walenga, J.M. (2000) Leuko-cyte activation in heparin-induced thrombo-cytopenia Blood 96, 29b.

19. Pouplard, C., Iochmann, S., Renard, B.,Herault, P., Colombat, P., Amiral, J., andGruel, Y. (2001) Induction of monocytetissue factor expression by antibodies toheparin-platelet factor 4 complexes devel-oped in heparin-induced thrombocytopeniaBlood 97, 3300–2.

20. Herbert, J.M., Savi, P., Jeske, W.P., andWalenga, J.M. (1998) Effect of SR121566A,a potent GP IIb-IIIa antagonist, on theHIT serum/heparin-induced platelet medi-ated activation of human endothelial cellsThromb Haemost 80, 326–31.

21. Walenga, J.M., Michal, K., Hoppensteadt,D., Wood, J.J., Bick, R.L., and Robinson,J.A. (1999) Vascular damage correlatesbetween heparin-induced thrombocytopeniaand the antiphospholipid syndrome ClinAppl Thromb Hemost 5(Suppl 1), S76–S84.

22. Fareed, J., Walenga, J.M., Hoppensteadt,D.A., Jeske, W.P., Lietz, H.W., Ahmad, S.,Callas, D., Messmore, H.L., and Haas, S.(1999) Selectins in the HIT syndrome:pathophysiologic role and therapeutic mod-ulation Semin Thromb Hemost 25(Suppl 1),37–42.

23. Blank, M., Shoenfeld, Y., Tavor, S.,Praprotnik, S., Boffa, M.C., Weksler, B.,Walenga, J.M., Amiral, J., and Eldor, A.(2002) Anti-platelet factor 4/heparin anti-bodies from patients with heparin-inducedthrombocytopenia provoke direct activa-tion of microvascular endothelial cells IntImmunol 14, 121–9.

24. Matsuo, T., Tomaru, T., Kario, K., andHirokawa, T. (2005) Incidence of heparin-PF4 complex antibody formation and

heparin-induced thrombocytopenia in acutecoronary syndrome Thromb Res 115,475–81.

25. Opatrny, L., and Warner, M.N. (2004) Riskof thrombosis in patients with malignancyand heparin-induced thrombocytopenia AmJ Hematol 76, 240–4.

26. Reininger, C.B., Greinacher, A., Graf, J.,Lassila, R., Steckmeier, B., and Schwieberer,L. (1996) Platelets of patients with peripheralarterial disease are hypersensitive to heparinThromb Res 81, 641–9.

27. Warkentin, T.E., Sheppard, J.A.,Horsewood, P., Simpson, P.J., and Moore,J.C. (2000) Impact of the patient populationon the risk for heparin-induced thrombocy-topenia Blood 96, 1703–8.

28. Shuster, T.A., Silliman, W.R., Coats, R.D.,Mureebe, L., and Silver, D. (2003) Heparin-induced thrombocytopenia: twenty-nineyears later J Vasc Surg 38, 1316–22.

29. Warkentin, T.E., and Greinacher, A. (2004)Heparin-induced thrombocytopenia: recog-nition, treatment, and prevention Chest 126,311S–37S.

30. Warkentin, T.E., and Crowther, M.A. (2007)When is HIT really HIT? Ann Thorac Surg83, 21–3.

31. Lo, G.K., Juhl, D., Warkentin, T.E., Sigouin,C.S., Eichler, P., and Greinacher, A. (2006)Evaluation of pretest clinical score (4 T’s) forthe diagnosis of heparin-induced thrombo-cytopenia in two clinical settings J ThrombHaemost 4, 759–65.

32. Pouplard, C., Amiral, J., Borg, J.Y., Laporte-Simitsidis, S., Delahousse, B., and Gruel, Y.(1999) Decision analysis for use of plateletaggregation test, carbon 14-serotonin releaseassay, and heparin-platelet factor 4 enzyme-linked immunosorbent assay for diagnosisof heparin-induced thrombocytopenia Am JClin Pathol 111, 700–6.

33. Arepally, G.M., and Ortel, T.L. (2006) Clin-ical practice. Heparin-induced thrombocy-topenia N Engl J Med 355, 809–17.

34. Parker, R.I. (2007) Measurement of heparin-dependent platelet antibodies in the diag-nosis of heparin-induced thrombocytopenia:fact or fiction? Crit Care Med 35, 1784–5.

35. Lillo-Le Louet, A., Boutouyrie, P., Alhenc-Gelas, M., Le Beller, C., Gautier, I.,Aiach, M., and Lasne, D. (2004) Diagnos-tic score for heparin induced thrombocytope-nia after cardiopulmonary bypass J ThrombHaemost 2, 1882–8.

36. Chan, M., Malynn, E., Shaz, B., and Uhl, L.(2008) Utility of consecutive repeat HITELISA testing for heparin-induced thrombo-cytopenia Am J Hematol 83, 212–7.


37. Izban, K.F., Lietz, J.W., Hoppensteadt, D.A.,Jeske, W.P., Fareed, J., Bakhos, M., andWalenga, J.M. (1999) Comparison of twoPF4/Heparin ELISA assays for the labora-tory diagnosis of heparin-induced thrombo-cytopenia Semin Thromb Hemost (Suppl 1)25, 51–6.

38. Warkentin, T.E., and Greinacher, A. (2007)Laboratory testing for heparin-inducedthrombocytopenia In: Warkentin, T.E.,and Greinacher, A., eds. Heparin-InducedThrombocytopenia, 4th edition. New York:Informa Healthcare; 2007:227–60.

39. Martin-Toutain, I., Piette, J.C., Diemert,M.C., Faucher, C., Jobic, L., and Ankri, A.(2007) High prevalence of antibodies toplatelet factor 4 heparin in patients withantiphospholipid antibodies in absence ofheparin-induced thrombocytopenia Lupus16, 79–83.

40. Warkentin, T.E., Sheppard, J.A., Moore,J.C., Moore, K.M., Sigouin, C.S., andKelton, J.G. (2005) Laboratory testing forthe antibodies that cause heparin-inducedthrombocytopenia: how much class to weneed? J Lab Clin Med 146, 341–6.

41. Juhl, D., Eichler, P., Lubenow, N.,Strobel, U., Wessel, A., and Greinacher, A.(2006) Incidence and clinical significanceof anti-PF4/heparin antibodies of the IgG,IgM, and IgA class in 755 consecutivepatients samples referred for diagnostic test-ing for heparin-induced thrombocytopeniaEur J Haematol 76, 420–6.

42. ISTH Platelet Immunology, Scientific Sub-committee. Chairs: Warkentin T, Chong B,Greinacher A, Gruel Y, Kiefel V, Kroll H.July 6, 2007, Palexpo, Geneva, Switzerland.Available at www.isth2007.com, Accessed onDec. 6, 2007.

43. Alberio, L., Kimmerle, S., Baumann, A.,Taleghani, B.M., Biasiutti, F.D., andLammle, B. (2003) Rapid determination ofanti-heparin/platelet factor 4 antibody titersin the diagnosis of heparin-induced throm-bocytopenia Am J Med 114, 528–36.

44. Francis, J.L. (2003) A critical evaluation ofassays for detecting antibodies to the heparin-PF4 complex Semin Thromb Hemost 30,359–68.

45. Francis, J.L., Drexler, A., Duncan, M.K.,Desai, H., Amaya, M., Robson, T., Meyer,T.V., Reyes, E., Rathmann, K., andAmirkhosravi, A. (2006) Prospective eval-uation of laboratory tests for the diagnosisof heparin-induced thrombocytopenia Blood108, 312a–3a.

46. Low, J., Austin, S., Bryant, A., and Joseph, J.(2007) Combining 4T’s pre-test clinical

score and ID-PAGIA heparin/PF4 test toexclude heparin induced thrombocytopenia(HIT) J Thromb Haemost (Suppl 2) 5,P-W-332.

47. Walenga, J.M., Jeske, W.P., Fasanella, A.R.,Wood, J.J., and Bakhos, M. (1999) Lab-oratory tests for the diagnosis of heparin-induced thrombocytopenia Semin ThrombHemost (Suppl 1) 25, 43–9.

48. Chong, B.H., Burgess, J., and Ismail, F.(1993) The clinical usefulness of the plateletaggregation test for the diagnosis of heparin-induced thrombocytopenia Thromb Haemost69, 344–50.

49. Warkentin, T.E., Hayward, C.P.M., Smith,C.A., Kelly, P.M., and Kelton, J.G. (1992)Determinants of donor platelet variabilitywhen testing for heparin-induced thrombo-cytopenia J Lab Clin Med 120, 371–9.

50. Eichler, P., Budde, U., Haas, S., Kroll, H.,Loreth, R.M., Meyer, O., Pachmann, U.,Potzsch, B., Schabel, A., Albrecht, D.,and Greinacher, A. (1999) First workshopfor detection of heparin-induced antibodies:validation of the heparin-induced platelet-activation test (HIPA) in comparison witha PF4/heparin ELISA Thromb Haemost 81,625–9.

51. Greinacher, A., Michels, I., Kiefel, V.,and Mueller-Eckhardt, C. (1991) A rapidand sensitive test for diagnosing heparin-induced thrombocytopenia Thromb Haemost66, 734–6.

52. Tomer, A. (1997) A sensitive and specificfunctional flow cytometric assay for the diag-nosis of heparin-induced thrombocytopeniaBr J Haematol 98, 648–56.

53. Tomer, A., Masalunga, C., and Abshire, T.C.(1999) Determination of heparin-inducedthrombocytopenia: a rapid flow cytometricassay for direct demonstration of antibody-mediated platelet activation Am J Hematol61, 53–61.

54. Stewart, M.W., Etches, W.S., Boshkov, L.K.,and Gordon, P.A. (1995) Heparin-inducedthrombocytopenia: an improved method ofdetection based on lumi-aggregomaetry Br JHaematol 91, 173–7.

55. Pouplard, C., Gueret, P., Fouassier, M.,Ternisien, C., Trossaert, M., and Regina, S.(2007) Prospective evaluation of the ‘4Ts’score and particle gel immunoassay spe-cific to heparin/PF4 for the diagnosis ofheparin-induced thrombocytopenia J ThrombHaemost 5, 1373–9.

56. Warkentin, T.E. (2002) Platelet count mon-itoring and laboratory testing for heparin-induced thrombocytopenia Arch Path LabMed 126, 1415–23.


57. Zwicker, J.I., Uhl, L., Huang, W.-Y., Shaz,B.H., and Bauer, K.A. (2004) Thrombosisand ELISA optical density values in hospital-ized patients with heparin-induced thrombo-cytopenia J Thromb Haemost 2, 2133–7.

58. Altunas, F., Matevosyan, K., Burner, J., Shen,Y.M., and Sarode, R. (2008) Higher opticaldensity of an antigen assay predicts thrombo-sis in patients with heparin-induced throm-bocytopenia Eur J Haematol 80, 429–35.

59. Refaai, M.A., Laposata, M., and VanCott,E.M. (2003) Clinical significance of a border-line titer in a negative ELISA test for heparin-induced thrombocytopenia Am J Clin Pathol119, 61–5.

60. Smythe, M.A., Koerber, J.M., and Mattson,J.C. (2005) An evaluation of heparin plateletfactor 4 antibody testing J Thromb Haemost(Suppl 1) 3, P1516.

61. Stribling, W.K., Slaughter, T.F., Houle, T.T.,and Sane, D.C. (2007) Beyond the plateletcount: heparin antibodies as independent riskpredictors Am Heart J 153, 900–6.

62. Mattioli, A.V., Bonetti, L., Sternieri, S., andMattioli, G. (2000) Heparin-induced throm-bocytopenia in patients treated with unfrac-tionated heparin: prevalence of thrombosis ina 1 year follow-up Ital Heart J 1, 39–42.

63. Williams, R.T., Damaraju, L.V., Mascelli,M.A., Barnathan, E.S., Califf, R.M.,Simoons, M.L., Deliargyris, E.N., and Sane,D.C. (2003) Anti-platelet factor 4/hep-arin antibodies: an independent predic-tor of 30 day myocardial infarction afteracute coronary syndromes Circulation 107,2307–12.

64. Bennett-Guerrero, E., Slaughter, T.F.,White, W.D., Welsby, I.J., Greenberg, C.S.,El-Moalem, H., and Ortel, T.L. (2005)Preoperative anti-PF4/heparin antibodylevel predicts adverse outcome after car-diac surgery J Thorac Cardiovasc Surg 130,1567–72.

65. Pena de la Vega, L., Miller, R.S., Benda,M.M., Grill, D.E., Johnson, M.G.,McCarthy, J.T., and McBane, R. (2005)Association of heparin-dependent antibod-ies and adverse outcomes in hemodialysispatients: a population-based study Mayo ClinProc 80, 995–1000.

66. Lindhoff-Last, E., Eicher, P., Stein, M.,Plagemann, J., Gerdsen, F., Wagner, R.,Ehrly, A.M., and Bauersachs, R. (2000)A prospective study on the incidence andclinical relevance of heparin-induced anti-bodies in patients after vascular surgeryThromb Res 97, 387–93.

67. Warkentin, T.E., Levine, M.N., Hirsch, J.,Horsewood, P., Roberts, R.S., Gent, M.,

and Kelton, J.G. (1995) Heparin-inducedthrombocytopenia in patients treated withlow-molecular-weight heparin or unfraction-ated heparin N Engl J Med 332, 1330–5.

68. Carrier, M., Rodger, M.A., Fergusson, D.,Doucette, S., Kovacs, M.J., Moore, J.,Kelton, J.G., and Knoll, G.A. (2008)Increased mortality in hemodialysis patientshaving specific antibodies to the platelet fac-tor 4-heparin complex Kidney Internat 73,213–9.

69. Wallis, D.E., Workman, D.L., Lewis, B.F.,Pifarré, R., and Moran, J.F. (1999) Fail-ure of early heparin cessation as a treatmentfor heparin-induced thrombocytopenia Am JMed 106, 629–35.

70. Messmore, H.L., Jeske, W.P., Wehrmacher,W.H., and Walenga, J.M. (2003) Benefit-riskassessment of treatments for heparin-inducedthrombocytopenia Drug Safety 26, 625–41.

71. Lewis, B.E., Wallis, D.E., Berkowitz, S.D.,Matthai, W.H., Fareed, J., Walenga, J.M.,Bartholomew, J., Sham, R., Lerner,R.G., Zeigler, Z.R., Rustagi, P.K., Jang, I.K.,Rifkin, S.D., Moran, J., Hursting, M.J.,Kelton, J.G., and for the ARG-911 StudyInvestigators (2001) Argatroban antico-agulant therapy in patients with heparin-induced thrombocytopenia Circulation 103,1838–43.

72. Lewis, B.E., Wallis, D.E., Leya, F., Hursting,M.J., Kelton, J.G., and for the Argatroban-915 Investigators (2003) Argatroban anti-coagulation in patients with heparin-inducedthrombocytopenia Arch Intern Med 163,1849–56.

73. Greinacher, A., Völpel, H., Janssens, U.,Hach-Wunderle, V., Kemkes-Matthes, B.,Eichler, P., Mueller-Velten, H.G.,Potzsch, B., and for the HIT InvestigatorsGroup (1999) Recombinant hirudin (lep-irudin) provides safe and effective antico-agulation in patients with heparin-inducedthrombocytopenia: a prospective study Cir-culation 99, 73–80.

74. Greinacher, A., Janssens, U., Berg, G.,Bock, M., Kwasny, H., Kemkes-Matthes,B., Eichler, P., Volpel, H., Potzsch, B.,Luz, M., and for the HAT Investigators(1999) Lepirudin (recombinant hirudin) forparenteral anticoagulation in patients withheparin-induced thrombocytopenia Circula-tion 100, 587–93.

75. Walenga, J.M., Fasanella, A.R., Iqbal, O.,Hoppensteadt, D.A., Ahmad, S., Wallis,D.E., and Bakhos, M. (1999) Coagula-tion laboratory testing in patients treatedwith argatroban Semin Thromb Hemost25(Suppl 1), 61–6.


76. Walenga, J.M., Drenth, A.F., Mayuga, M.,Hoppensteadt, D.A., Prechel, M.M.,Harder, S., Watanabe, H., Osakabe, M., andBreddin, H.K. (2008) Transition from arga-troban to oral anticoagulation with phen-procoumon or acenocoumarol: effect oncoagulation factor testing Clin Appl ThrombHemost 14, 325–31.

77. Hirsh, J., Heddle, N., and Kelton, J.G.(2004) Treatment of heparin-inducedthrombocytopenia Arch Intern Med 164,361–9.

78. Warkentin, T.E. (2003) Management ofheparin-induced thrombocytopenia: a criti-cal comparison of lepirudin and argatrobanThromb Res 110, 73–82.

79. LeMonte, M.P., Brown, P.M., and Hurst-ing, M.J. (2005) Alternative parenteral anti-coagulation with argatroban, a direct throm-bin inhibitor Expert Rev Cardiovasc Ther 3,31–41.

80. Lewis, B.E., Walenga, J.M., and Hursting,M.J. (2002) Argatroban anticoagulation inpatients with heparin-induced thrombocy-topenia Cardiovasc Rev Rep 23, 445–57.

81. Swan, S.K., and Hursting MJ. (2000) Thepharmacokinetics and pharmacodynamics ofargatroban: effects of age, gender, and hep-atic or renal dysfunction Pharmacotherapy20, 318–29.

82. Swan, S.K., St. Peter, J.V., Lambrecht, L.J.,and Hursting, M.J. (2000) Comparison ofanticoagulant effects and safety of argatrobanand heparin in healthy subjects Pharma-cotherapy 20, 756–70.

83. Matthai, W.H., Jr., Hursting, M.J., Lewis,B.E., and Kelton, J.G. (2005) Argatrobananticoagulation in patients with a history ofheparin-induced thrombocytopenia ThrombRes 116, 121–6.

84. Walenga, J.M., Ahmad, S., Hoppensteadt,D., Iqbal, O., Hursting, M.J., and Lewis,B.E. (2002) Argatroban therapy does notgenerate antibodies that alter its antico-agulant activity in patients with heparin-induced thrombocytopenia Thromb Res 105,401–5.

85. Walenga, J.M., Fareed, D., Schultz, C.,Neville, B., and Hoppensteadt, D. (2004)Argatroban, not lepirudin or bivalirudin,treatment results in the generation of nitricoxide during parenteral administration Blood104, 512a.

86. Lewis, B.E., Matthai, W.H., Cohen, M.,Moses, J., Hursting, J., and Leya F. (2002)Argatroban anticoagulation during percuta-neous coronary intervention in patients withheparin-induced thrombocytopenia CathetCardiovasc Intervent 57, 177–84.

87. Cornell, T., Wyrick, P., Fleming, G.,Pasko, D., Han, Y., Custer, J., Haft, J., andAnnich, G. (2007) A case series describingthe use of argatroban in patients on extracor-poreal circulation ASAIO J 53, 460–3.

88. Potter, K.E., Raj, A., and Sullivan, J.E.(2007) Argatroban for anticoagulation inpediatric patients with heparin-inducedthrombocytopenia requiring extracorporeallife support J Ped Hematol Oncol 29, 265–8.

89. Jang, I., and Hurstings, M.J. (2005) Whenheparins promote thrombosis: review ofheparin-induced thrombocytopenia Circula-tion 111, 2671–83.

90. Cetta, F., Graham, L.C., Wrona, L.L.,Arruda, M.J., and Walenga, J.M. (2004)Argatroban use during pediatric interven-tional cardiac catheterization Cathet Cardio-vasc Intervent 61, 147–9.

91. Jang, I.K., Lewis, B.E., Matthai, W.H., andKleiman, N.S. (2004) Argatroban antico-agulation in conjunction with glycoproteinIIb/IIIa inhibition in patients undergo-ing percutaneous coronary intervention:an open-label, nonrandomized pilot studyJ Thromb Thrombol 18, 31–7.

92. LaMonte, M.P., Brown, P.M., and Hursting,M.J. (2004) Stroke in patients with heparin-induced thrombocytopenia and the effectof argatroban therapy Crit Care Med 32,976–80.

93. Eichler, P., Friesen, H.J., Lubenow, N.,Jaeger, B., and Greinacher, A. (2000)Anti-hirudin antibodies in patients withheparin-induced thrombocytopenia treatedwith lepirudin: incidence, effects on aPTT,and clinical relevance Blood 96, 2373–8.

94. Greinacher, A., Eichler, P., Albrecht, D.,Strobel, U., Potzsch, B., and Eriksson,B.I. (2003) Anti-hirudin antibodies follow-ing low-dose subcutaneous treatment withdesirudin for thrombosis prophylaxis afterhip-replacement surgery: incidence and clini-cal relevance Blood 101, 2617–9.

95. Fischer, K.G., Liebe, V., Hudek, R.,Piazolo, L., Haase, K.K., Borggrefe, M., andHuhle, G. (2003) Anti-hirudin antibodiesalter pharmacokinetics and pharmacodynam-ics of recombinant hirudin Thromb Haemost89, 973–82.

96. Greinacher, A., Lubenow, N., and Eichler, P.(2003) Anaphylactic and anaphylactoid reac-tions associated with lepirudin in patientswith heparin-induced thrombocytopenia(HIT) Circulation 108, 2062–5.

97. Cochran, K., DeMartini, T.J., Lewis, B.E.,O’Brien, J., Steen, L.H., Grassman, E.D.,and Leya, F. (2003) Use of lepirudin dur-ing percutaneous vascular interventions in


patients with heparin-induced thrombocy-topenia J Invas Cardiol 15, 622–3.

98. Mahaffey, K.W., Lewis, B.E., Wildermann,N.M., Berkowitz, S.D., Oliverio, R.M.,Turco, M.A., Shalev, Y., Lee, P.V., Traverse,J.H., Rodriguez, A.R., Ohman, E.M.,Harrington, R.A., Califf, R., and for theATBAT Investigators (2003) The anticoag-ulant therapy with bivalirudin to assist in theperformance of PCI in patients with heparin-induced thrombocytopenia (ATBAT) studyJ Invas Cardiol 15, 611–6.

99. Merry, A.F., Raudkivi, P.J., Middleton,N.G., McDougall, J.M., Nand, P., Mills,B.P., Webber, B.J., Frampton, C.M., andWhite, H.D. (2004) Bivalirudin versus hep-arin and protamine in off-pump coronaryartery bypass surgery Ann Thorac Surg 77,925–31.

100. Koster, A., Spiess, B., Chew, D.P.,Krabatsch, T., Tambeur, L., DeAnda, A.,Hetzer, R., Kuppe, H., Smedira, N.G.,and Lincoff, A.M. (2004) Effectiveness ofbivalirudin as a replacement for heparinduring cardiopulmonary bypass in patientsundergoing coronary artery bypass graftingAm J Cardiol 93, 356–9.

101. Magnani, H.N., and Gallus, A. (2006)Heparin-induced thrombocytopenia (HIT).A report of 1,478 clinical outcomes ofpatients treated with danaparoid (Orgaran)from 1982 to mid-2004 Thromb Haemost95, 967–81.

102. Lindhoff-Last, E., Kreutzenbeck, H-J., andMagnani H.N. (2005) Treatment of 51 preg-nancies with danaparoid because of heparinintolerance Thromb Haemost 93, 63–9.

103. Chong, B.H., Gallus, A.S., Cade, J.F.,Magnani, H., Manoharan, A., Oldmeadow,M., and for the Australian HIT Study Group(2001) Prospective randomized open-labelcomparison of danaparoid with dextran 70 inthe treatment of heparin-induced thrombo-cytopenia with thrombosis Thromb Haemost86, 1170–5.

104. Farner, B., Eichler, P., Kroll, H., andGreinacher, A. (2001) A comparison ofdanaparoid and lepirudin in heparin-inducedthrombocytopenia Thromb Haemost 85,950–7.

105. Walenga, J.M., Koza, M.J., Lewis, B.E., andPifarré, R. (1996) Relative heparin-inducedthrombocytopenic potential of low molecu-lar weight heparins and new antithromboticagents Clin Appl Thromb Hemost 2(Suppl 1),S21–7.

106. Elalamy, I., Lecrubier, C., Horellou,M.H., Conard, J., and Samama, M.M.(2000) Heparin-induced thrombocytopenia:

laboratory management Ann Med32(Suppl 1), 60–7.

107. Parody, R., Oliver, A., Souto, J.C., andFontcuberta, J. (2003) Fondaparinux(Arixtra) as an alternative anti-thromboticprophylaxis when there is hypersensitivityto low molecular weight and unfractionatedheparins Haematologica 88, ECR32.

108. D’Amico, E.A., Villaca, P.R., Gualandro,S.F., Bassitt, R.P., and Chamone, D.A.(2003) Successful use of Arixtra in a patientwith paroxysmal nocturnal hemoglobinuria,Budd-Chiari syndrome and heparin-inducedthrombocytopenia J Thromb Haemost 1,2452–3.

109. Bradner, J., Hallisey, R.K., and Kuter,D.J. (2004) Fondaparinux in the treatmentof heparin-induced thrombocytopenia Blood104, Abstract 1775.

110. Lobo, B., Finch, C., Howard, A., andMinhas S. (2008) Fondaparinux for thetreatment of patients with acute heparin-induced thrombocytopenia Thromb Haemost99, 208–14.

111. Warkentin TE, Maurer, B.T., and Aster, R.H.(2007) Heparin-induced thrombocytopeniaassociated with fondaparinux N Engl J Med356, 2653–5.

112. Rota, E., Bazzan, M., and Fantino, G.(2008) Fondaparinux-related thrombocy-topenia in a previous low-molecular-weightheparin (LMWH)-induced heparin-inducedthrombocytopenia (HIT) Thromb Haemost99, 779–81.

113. Warkentin, T.E., Cook, R.J., Marder, V.J.,Sheppard, J.A., Moore, J.C., Eriksson, B.I.,Greinacher, A., and Kelton, J.G. (2005)Anti-platelet factor 4/heparin antibodiesin orthopedic surgery patients receivingantithrombotic prophylaxis with fonda-parinux or enoxaparin Blood 106, 3791–6.

114. Eriksson, B.I., Borris, L.C., Dahl, O.E.,Haas, S., Huisman, M.V., Kakkar, A.K.,Misselwitz, F., Muehlhofer, E., andKalebo, P. (2007) Dose-escalation study ofrivaroxaban (BAY 59-7939), an oral, directfactor Xa inhibitor, for the prevention ofvenous thromboembolism in patients under-going total hip replacement Thromb Res 120,685–93.

115. Agnelli, G., Gallus, A., Goldhaber, S.Z.,Haas, S., Huisman, M.V., Hull, R.D.,Kakkar, A.K., Misselwitz, F., Schellong, S.,and for the ODIXa-DVT Study Inves-tigators. (2007) Treatment of proximaldeep-vein thrombosis with the oral directfactor Xa inhibitor rivaroxaban (BAY 59-7939): the ODIXa-DVT (Oral direct fac-tor Xa inhibitor BAY 59-7939 in patients


with acute symptomatic deep-vein thrombo-sis) study Circulation 116, 180–7.

116. Erikkson, B.I., Borris, L.C., Dahl, O.E.,Haas, S., Huisman, M.V., Kakkar, A.K.,Muehlhofer, E., Dierig, C., Misselwitz, F.,Kalebo, P., and for the ODIXa-HIP StudyInvestigators. (2006) A once-daily, oral directfactor Xa inhibitor, rivaroxaban (BAY 59-7939), for thromboprophylaxis after total hipreplacement Circulation 114, 2374–81.

117. Lassen, M.R., Davidson, B.L., Gallus, A.,Pineo, G., Ansell, J., and Deitchman, D.(2007) The efficacy and safety of apixban, anoral, direct factor Xa inhibitor, as thrombo-prophylaxis in patients following total kneereplacement J Thromb Haemost 5, 2368–75.

118. Walenga, J.M., Jeske, W.P., Prechel, M.,Hoppensteadt, D., Swank, J., Sheen, L.,Misselwitz, F., Messmore, H., andBakhos, M. (2007) Potential of the fac-tor Xa inhibitor rivaroxaban for the anti-coagulation management of patients withheparin-induced thrombocytopenia J ThrombHaemost 5(Suppl 2), P-M-648.

119. Wallis, D.E., Quintos, R., Wehrmacher, W.,and Messmore, H.L. (1999) Safety ofwarfarin anticoagulation in patients withheparin-induced thrombocytopenia Chest116, 1333–8.

120. Gosselin, R.C., Dager, W.E., King, J.H.,Janatpour, K., Mahackian, K., Larkin, E.C.,and Owings, J.T. (2004) Effect of directthrombin inhibitors, bivalirudin, lepirudin,and argatroban, on prothrombin time andINR values Am J Clin Pathol 121, 593–9.

121. Sheth, S.B., DiCicco, R.A., Hursting, M.J.,Montague, T., and Jorkasky, D.K. (2001)Interpreting the international normalizedratio (INR) in individuals receiving arga-troban and warfarin Thromb Haemost 85,435–40.

122. Harder, S., Graff, J., Klinkhardt, U.,von Hentig, N., Walenga, J.M., Watanabe,H., Osakabe, M., and Breddin, H.K. (2004)Transition from argatroban to oral antico-agulation with phenprocoumon or aceno-coumarol: effects on PT, aPTT, and ecarinclotting time Thromb Haemost 91, 1137–45.

123. Hursting, M.J., Lewis, B.E., and Macfarlane,D.E. (2005) Transitioning from arga-troban to warfarin therapy in patients withheparin-induced thrombocytopenia ClinAppl Thromb Hemost 11, 279–87.

124. Pötzsch, B., and Klovekorn, W.P. (2000) Useof heparin during cardiopulmonary bypass inpatients with a history of heparin-induced

thrombocytopenia N Engl J Med 343,515–6.

125. Walenga, J.M., Jeske, W.P., Wallis, D.E.,Bakhos, M., Lewis, B.E., Leya, F., andFareed, J. (1999) Clinical experience withcombined treatment of thrombin inhibitorsand GPIIb/IIIa inhibitors in patients withHIT Semin Thromb Hemost 25(Suppl 1),77–81.

126. Walenga, J.M., Lewis, B.E., Jeske, W.P.,Leya, F., Wallis, D.E., Bakhos, M., andFareed, J. (1999) Combined thrombin andplatelet inhibition treatment for HIT patientsHämostaseologie 19, 128–33.

127. Haas, S., Walenga, J.M., Jeske, W.P., andFareed, J. (1999) Heparin-induced throm-bocytopenia: the role of platelet activationand therapeutic implications Semin ThrombHemost 25(Suppl 1), 67–75.

128. Fareed, J., Bick, R.L., Rao, G., Goldhaber,S.Z., Sasahara, A., Messmore, H.L.,Hoppensteadt, D.A., and Nicolaides, A.(2008) The immunogenic potential ofgeneric versions of low-molecular weightheparins may not be the same as the brandedproducts Clin Appl Thromb Hemost 14, 5–7.

129. Houiste, C., Auguste, C., Macrez, C.,Dereux, S., Derouet, A., and Anger, P.(2009) Quantitative PCR and disaccharideprofiling to characterize the animal originof low-molecular-weight heparins Clin ApplThromb Hemost 15, 50–8.

130. Kishimoto, T.K., Viswanathan, K.,Ganguly, T., Elankumaran, S., Smith, S.,Pelzer, K., Lansing, J.C., Sriranganathan, N.,Zhao, G., Galcheva-Gargova, Z., Al-Hakim,A., Bailey, G.S., Fraser, B., Roy, S., Rogers-Cotrone, T., Buhse, L., Whary, M., Fox,J., Nasr, M., Dal Pan, G.J., Shriver, Z.,Langer, R.S., Venkataraman, G., Austen,K.F., Woodcock, J., and Sasisekharan, R.(2008) Contaminated heparin associatedwith adverse clinical events and activationof the contact system N Engl J Med 358,2457–67.

131. Guerrini, M., Beccati, D., Shriver, Z.,Naggi, A., Viswanathan, K., Bisio, A., Capila,I., Lansing, J.C., Guglieri, S., Fraser,B., Al-Hakim, A., Gunay, N.S., Zhang,Z., Robinson, L., Buhse, L., Nasr, M.,Woodcock, J., Langer, R., Venkataraman,G., Linhardt, R.J., Casu, B., Torri, G., andSasisekharan, R. (2008) Oversulfated chon-droitin sulfate is a contaminant in heparinassociated with adverse clinical events NatBiotechnol 26, 669–75.

Chapter 5

Novel Anticoagulant Therapy: Principle and Practice

Shaker A. Mousa

Abstract

Currently, there are several lines of evidence supporting the interplay between coagulation andinflammation in the propagation of various disease processes, including venous thromboembolism (VTE)and inflammatory diseases. Major advances in the development of oral anticoagulants have resulted inconsiderable progress toward the goal of safe and effective oral anticoagulants that do not require fre-quent monitoring or dose adjustment and have minimal food/drug interactions. Indirect inhibitors suchas low-molecular-weight heparin (LMWH) and the pentasaccharide fondaparinux represent improve-ments over traditional drugs such as unfractionated heparin for acute treatment of VTE, constituting amore targeted anticoagulant approach with predictable pharmacokinetic profiles and no requirement formonitoring. Vitamin K antagonist, with its inherent limitations in terms of multiple food and drug inter-actions and frequent need for monitoring, remains the only oral anticoagulant approved for long-termsecondary thromboprophylaxis in VTE. The oral-direct thrombin inhibitor ximelagatran was withdrawnfrom the world market due to safety concerns. Newer anticoagulant drugs such as parenteral pentasac-charides (idraparinux, SSR126517E), novel oral-direct thrombin inhibitors (dabigatran), oral-direct fac-tor Xa inhibitors (rivaroxaban, apixaban, YM-150, DU-176b), and tissue factor/factor VIIa complexinhibitors have been “tailor-made” to target specific procoagulant complexes and have the potential togreatly expand oral antithrombotic targets for both acute and long-term treatment of VTE, acute coro-nary syndromes, and for the prevention of stroke in atrial fibrillation patients.

Key words: Anticoagulants, direct thrombin inhibitors, factor Xa inhibitors, factor IXa inhibitors,pentasaccharide, tissue factor/factor VIIa complex inhibitors, venous thromboembolism.

1. Introduction

Inflammation plays a key role in triggering a prothrombotic statethrough the activation of platelets, induction of coagulation, andvascular insult. Venous thromboembolism (VTE) continues to bea major cause of morbidity and mortality in the western world


157

158 Mousa

(1). VTE represents 1 in 10 hospital deaths, and post-thromboticsyndrome and pulmonary hypertension occur in 10% of deep-veinthrombosis (DVT) and 5% of pulmonary embolism (PE) patients,respectively (2). For more than 50 years, traditional drugs, suchas unfractionated heparin (UFH), have been used parenterally foracute treatment, followed by oral vitamin K antagonists (VKAs)such as warfarin for long-term treatment. These drugs exerttheir antithrombotic effects by inhibiting multiple steps of thecoagulation cascade, but there are inherent limitations for eachdrug.

For acute VTE treatment, the limitations of UFH include aless than predictable anticoagulant response with the need forfrequent monitoring, and the potential for severe toxicity, espe-cially heparin-induced thrombocytopenia (HIT) in up to 3% ofpatients (3). Over the past 15 years, the use of LMWH and morerecently the synthetically derived pentasaccharide fondaparinuxhas improved the acute management of VTE. A more targetedapproach to procoagulant complex inhibition, predictable phar-macodynamic characteristics, and improved safety profiles haveenabled complete treatment of VTE on an outpatient basis forselect patients without the need for anticoagulant monitoring.Other parenteral drugs such as the direct thrombin inhibitors(DTIs) lepirudin and argatroban have achieved only limited use inacute VTE treatment, namely in thrombosis associated with HIT.

Optimal long-term treatment of VTE is defined by the limi-tations of VKAs, the only oral anticoagulants currently approvedfor use in this setting. These limitations include a slow onset ofaction and the need for bridging anticoagulation with a parenteraldrug in the acute setting, multiple food and drug interactions,and a narrow therapeutic window, necessitating frequent coag-ulation monitoring and dose adjustment (4). In addition, somepatient subgroups cannot tolerate VKA, such as pregnant patientsrequiring anticoagulation, in whom VKA is associated with a riskof teratogenicity (5), or patients in whom VKA is associated withhigher risks of recurrent thromboembolism and major bleeding,such as those with active cancer (6, 7). In both of these patientgroups, emerging data support the use of long-term LMWH (8–10), with limited parenteral use.

An improved understanding of the molecular mechanisms ofcoagulation and thrombosis and the potential to apply this knowl-edge at the clinical level to different patient subgroups has ledto the development of newer antithrombotic drugs for use inVTE treatment. Many of these drugs are orally active, syntheti-cally derived, and target-specific procoagulant complexes withinthe coagulation cascade (11). These drugs can broadly be cat-egorized as interfering with the initiation of coagulation [tissuefactor/factor VIIa (TF/FVIIa) complex], propagation of coagu-lation [indirect and direct inhibitors of activated factor X (FXa)

Novel Anticoagulant Therapy: Principle and Practice 159

or FIXa], and thrombin activity (DTIs). This chapter will focuson these types of investigational drugs for VTE treatment, withan emphasis on those undergoing or that have recently completedphase-II or III clinical trials (Table 5.1). The effect of anticoagu-lants on stroke in atrial fibrillation patients will also be discussed.Atrial fibrillation (AF) is an epidemic, affecting 11.5% of the pop-ulation in the developed world. Its significance lies predominantlyin that AF patients have a fivefold increased risk of stroke. Strokesassociated with AF are usually more severe and confer increasedrisk of morbidity and mortality and poor functional outcomes.Despite advances in the development of promising experimentalapproaches for select patients with acute stroke, primary preven-tion with pharmacological agents remains the best approach toreducing the burden of stroke.

2. AnticoagulantTargets ThatInhibit theInitiation Phase

2.1. TF/FVIIaComplex Inhibitors

The TF/FVIIa complex, as part of the extrinsic system of thecoagulation cascade, is considered to be the key system for theinitiation of coagulation. In the venous vascular system, expo-sure of TF in orthopedic surgery and in subsets of cancer patients(12, 13) is believed to be responsible for the high risk of VTEin these patient groups, making pharmacological inhibition ofthe TF/FVIIa complex important (14). The function of TFcan be blocked using several approaches, including antibodiesthat prevent the binding of FVIIa to TF, active site inhibitionof FVIIa, small molecules or antibodies that block TF/FVIIacomplex function, and molecules that inhibit the active site ofFVIIa in the TF/FVIIa complex after binding to FXa (15, 16).Moreover, TF pathway inhibitor (TFPI), a naturally occurringinhibitor, forms a neutralizing complex with TF/FVIIa and FXa(14). Thus TFPI, either by upregulation of endogenous TFPIpools or exogenous administration of recombinant TFPI (rTFPI),represents an attractive anticoagulant that acts by blocking thepathological impact of TF/VIIa and Xa on coagulation andbeyond (17).

2.2. NematodeAnticoagulantProteins

The hematophagous hookworm Ancylostoma caninum producesa family of small, disulfide-linked protein anticoagulants (75–84amino acid residues) referred to as nematode anticoagulant pro-teins (NAPs), which have been the focus of antithrombotic drugdevelopment efforts due to their ability to inhibit the TF/FVIIa

160 Mousa

Tabl

e5.

1Ex

ampl

esof

oral

antic

oagu

lant

sin

phas

e-II/

IIItr

ials

Drug

aTa

rget

Dosi

ngAd

just

men

tan

din

tera

ctio

nsIn

dica

tions

Riv

arox

aban

(Bay

erSc

heri

ng)

Fact

orX

aO

nce

ortw

ice

daily

•A

void

orm

onito

rw

ithC

rCl<

30m

l/m

in•D

ose

redu

ctio

nfo

rCrC

l30

–49

ml/

min

•A

void

orm

onito

rw

ithst

rong

CYP

450

3A4

inhi

bito

rs

•Pha

seII

Ifo

rV

TE

prop

hyla

xis

afte

rjo

int

repl

acem

ent

com

plet

ed(n

=2,5

31)

•Pha

seII

for

seco

ndar

ypr

even

tion

afte

rA

CS

enro

lling

(n=1

,350

)•

Phas

eII

for

trea

tmen

tof

DV

Tco

mpl

eted

,ph

ase-

III

enro

lling

(n=2

,900

)•

Phas

eII

Ifo

rpr

even

tion

ofst

roke

inno

n-va

lvul

arA

Fen

rolli

ng(n

=14,

000)

•Pha

seII

Ifo

rtr

eatm

ent

ofPE

enro

lling

(n=3

,300

)

Api

xaba

n(B

rist

ol-M

yers

Squi

bban

dPfi

zer)

Fact

orX

aT

wic

eda

ily•

Avo

idor

mon

itor

with

stro

ngC

YP45

03A

4in

hibi

tors

•Pha

seII

tria

lfor

trea

tmen

tof

acut

eD

VT

com

plet

ed•P

hase

III

tria

lsfo

rV

TE

prev

entio

naf

ter

join

tre

plac

e-m

ent

enro

lling

(n=7

,258

)•P

hase

III

tria

lfor

prev

entio

nof

VT

Ein

med

ical

patie

nts

(n=6

,524

)•P

hase

III

tria

lfor

prev

entio

nof

stro

kein

non-

valv

ular

AF

enro

lling

(n=1

5,00

0)L

Y517

717

(Eli

Lill

y),

Fact

orX

aO

nce

daily

•Und

eter

min

edye

t•P

hase

IItr

ialf

orV

TE

prop

hyla

xisa

fter

join

trep

lace

men

tco

mpl

eted

Dab

igat

ran

(Boe

hrin

gerI

ngel

heim

)Fa

ctor

IIa

(thr

ombi

n)O

nce

ortw

ice

daily

•A

void

orm

onito

rw

ithpr

oton

pum

pin

hibi

tors

•Pha

seII

Itr

ialf

orV

TE

prev

entio

naf

terj

oint

repl

acem

ent

com

plet

ed•P

hase

III

tria

lfor

trea

tmen

tofV

TE

enro

lling

(n=2

,554

)•P

hase

III

tria

lfor

prev

entio

nof

stro

kein

non

valv

ular

AF

enro

lling

(n=1

5,00

0)

a Add

ition

alor

alan

ticoa

gula

nts

inpr

eclin

ical

orea

rly

clin

ical

deve

lopm

ents

are

not

liste

d.


complex. One of these nematode anticoagulant proteins, NAP5,inhibits the amidolytic activity of FXa with a Ki = 43 pM andis the most potent natural FXa inhibitor identified to date. NAP5does not inhibit FVIIa in the TF/FVIIa complex. Rather, it eitherbinds to FXa alone, or, as is the case for its family member NAPc2,in combination with a protein exosite, resulting in potent inhi-bition of the TF/FVIIa complex. NAPc2 has been tested sub-cutaneously for VTE prophylaxis in a phase I/II clinical trialusing mandatory unilateral venography in 293 patients undergo-ing total knee replacement surgery. At a dose of 3 μg/kg adminis-tered within 1 h after surgery, NAPc2 was associated with an over-all DVT rate of 12.2%, a proximal DVT rate of 1.3%, and a majorbleed rate of 2.3% (18). Further clinical trials with NAPc2 are inprogress (19, 20). Recombinant NAPc2, like other inhibitors ofTF/FVIIa, including TFPI and active site-blocked FVIIa (ASIS,FFR-rFVIIa, or FVIIai) have a promising role in the preventionand treatment of venous and arterial thrombosis and could poten-tially be efficacious in the management of disseminated intravas-cular coagulopathies due to their ability to potentently and selec-tively inhibit TF/FVIIa (19, 20).

2.3. Anti-TF/VIIa FVIIa is a key serine protease involved in the initiation of thecoagulation cascade. It is a glycosylated disulfide-linked het-erodimer comprised of a heavy chain and a light chain. Thelight chain contains an amino-terminal gamma-carboxyglutamicacid-rich (Gla) domain and two epidermal growth factor (EGF)-like domains, and there is a chymotrypsin-like serine proteasedomain in the heavy chain (21). TF, a membrane bound pro-tein, is an essential cofactor of FVIIa that is required for maxi-mal activity toward its biological substrates (FX, FIX, and FFVII).As such, the TF/FVIIa complex plays an important role in nor-mal physiology as well as in thrombotic diseases such as unstableangina (UA), disseminated intravascular coagulation (DIC), andDVT. In addition to its function as an initiator of coagulation,TF/FVIIa plays an important role in inflammation and angiogen-esis (22, 23). A wide array of strategic approaches to inhibiting thebiochemical and biological functions of the TF/FVIIa complexhave been pursued. These have been greatly enhanced by elucida-tion of the structures of TF, FVII, FVIIa, and the TF/FVIIa com-plex, resulting in inhibitors that are directed specifically towardeither FVIIa or TF. Antagonists of the TF/FIIa complex includeactive site-inhibited FVIIa, TF mutants, anti-TF antibodies, anti-FVII/FVIIa antibodies, naturally occurring protein inhibitors,peptide exosite inhibitors, and protein and small-molecule activesite inhibitors. These antagonists can inhibit catalysis directly atthe active site as well as impair function by binding to exositesthat interfere with substrate, membrane, or cofactor binding. Sev-eral different small-molecule potent inhibitors of TF/FVIIa have

162 Mousa

been shown to reduce thrombus weight in animal models anddecrease the level of interleukin-6 (IL-6) in a LPS-stimulatedmouse model of endotoxemia (22–25). A study designed to eval-uate the antithrombotic efficacy and bleeding propensity of aselective, small-molecule inhibitor of TF/VIIa in comparison tosmall-molecule, selective inhibitors of FXa and thrombin in anonhuman primate model of thrombosis was reported by Suley-manov et al. (25). The data indicated that TF/VIIa inhibi-tion effectively prevents arterial thrombosis, with less impact onbleeding parameters than equivalent doses of FXa and thrombininhibitors (25).

2.4. TFPI The anticoagulant effects of TFPI in animals (26, 27) and therole of TFPI release in secondary anticoagulant mechanisms ofLMWH action in humans (28) have been demonstrated. To date,rTFPI has been tested only in experimental models. TFPI is a nat-ural (i.e., endogenous) inhibitor of TF coagulant and signalingactivities. TFPI exerts anti-angiogenic and antimetastatic effectsin vitro and in vivo (17, 29). In animal models of experimen-tal metastasis, circulating and tumor cell-associated TFPI signifi-cantly reduce tumor cell-induced coagulation activation and lungmetastasis (29). Heparins and heparin derivatives, which inducethe release of TFPI from the vascular endothelium, also exhibitantitumor effects, and TFPI likely contributes significantly tothese effects (17). Recently, a non-anticoagulant LMWH withintact TFPI-releasing capacity was shown to have significantantimetastatic effects in an experimental mouse model (30). Evi-dence of dual inhibitory functions of TFPI on TF-driven coagula-tion and signaling strengthen the rationale for considering TFPIa potential anticancer agent (26).

TFPI-2, a member of the Kunitz-type serine proteinaseinhibitor family, is a structural homologue of TFPI. The expres-sion of TFPI-2 in tumors is inversely related to the degree ofmalignancy, suggesting a role for TFPI-2 in the maintenance oftumor stability and inhibition of growth of neoplasms (31). TFPI-2 inhibits the TF/VIIa complex and a wide variety of serineproteinases, including plasmin, plasma kallikrein, FXIa, trypsin,and chymotrypsin. Aberrant methylation of TFPI-2 promotercytosine-phosphorothioate-guanine (CpG) islands in human can-cers and cancer cell lines results in decreased expression of TFPI-2mRNA and decreased synthesis of TFPI-2 protein during cancerprogression (31). TFPI-2 has been shown to induce apoptosis andinhibit angiogenesis, thereby potentially contributing significantlyto inhibition of tumor growth. Restoration of TFPI-2 expressionin tumor tissue inhibits invasion, tumor growth, and metasta-sis, lending support to the use of TFPI-2-targeted therapeuticsas novel treatments for cancer (31).


3. Inhibitorsof CoagulationPropagation

It is widely accepted that FXa, as part of a prothrombinasecomplex with FVa, plays a central role in clot formation, giventhat it is generated by the extrinsic and intrinsic pathways ofcoagulation as they converge into the final common pathway.In addition, within the prothrombinase complex, one moleculeof FXa can exponentially generate 138 molecules of thrombinper minute. Theoretically, FXa inhibitors might have an advan-tage over thrombin inhibitors by preventing the activation ofcoagulation amplification mechanisms as well as thrombin gen-eration in both platelet-rich arterial thrombosis and fibrin-richvenous thrombosis, making FXa a prime target for anticoagulantdrug design. However, in animal models of thrombosis, selec-tive FXa inhibition was less potent than direct thrombin inhibi-tion in arterial and venous models of thrombosis, and associatedwith increased global clotting times, indicating that inhibition ofFXa is less effective than direct thrombin inhibition in controllingthrombin formation (32). The possibility that selective inhibitionof FXa upstream could result in a safer bleeding profile cannot beentirely ruled out; in the absence of thrombin inhibition, smallamounts of thrombin would escape neutralization and facilitatehemostasis.

Inhibitors of FXa include both indirect (antithrombin-mediated) and direct antithrombin (AT)-independent selectiveinhibitors. Other possible targets of coagulation propagation,through the prothrombinase complex or other routes, includeFIXa inhibitors, FVIIIa and FVa inhibitors, activated protein C,or soluble thrombomodulin.

3.1. Indirect FXaInhibitors

The synthetically derived pentasaccharides fondaparinux and idra-parinux represent the most advanced selective indirect FXainhibitors. These agents exert their action through high-affinitybinding and activation of AT, which then inhibits free FXa. Fon-daparinux contains the pentasaccharide sequence of heparin andselectively binds to and induces a conformational change in AT,increasing the anti-Xa activity of AT nearly 300-fold in a cat-alytic fashion. Fondaparinux has a linear pharmacokinetic profileand predictable anticoagulant response, with a plasma half-life ofapproximately 18 h and >95% bioavailability after intravenous orsubcutaneous injection, allowing non-monitored once-daily sub-cutaneous dosing. In addition, it does not bind to platelet factor 4(PF4) and has not been associated with drug-induced thrombocy-topenia. Fondaparinux is currently approved for acute treatmentof DVT and PE based on the recently completed MATISSE stud-ies in VTE (33, 34). As such, fondaparinux represents the first

164 Mousa

of a class of selective indirect FXa inhibitors to provide proof ofconcept that FXa inhibitors can be used to treat thrombosis inthe acute stage as efficaciously as drugs with established AT activ-ity. Fondaparinux is also the first of a new class of antithromboticdrugs designed specifically to inhibit a single target or procoagu-lant complex in the coagulation cascade.

3.1.1. Idraparinux(Sanofi-Aventis)

Idraparinux is a hypermethylated, long-acting pentasaccharidethat can be administered with once-weekly dosing. Idraparinuxsodium is a second-generation pentasaccharide with sulfated sidechains, which results in a 30-fold higher binding affinity to ATas compared to fondaparinux, and a 120-h elimination half-life,allowing once-weekly administration (35). It has similar clini-cal properties as fondaparinux, with 100% bioavailability afterparenteral administration, linear pharmacokinetics, a predictableanticoagulant response with no need for monitoring, does notinduce platelet aggregation, has no effects on PF4, and no evi-dence of induction of thrombocytopenia. The major drawback,as for fondaparinux, has been the lack of an antidote, althoughthe importance of an antidote in clinical practice is controversial.However, it should be noted that biotinylated idraparinux, dis-cussed in detail later, does have an antidote.

The PERSIST study, a randomized, phase-II, dose-rangingstudy, compared idraparinux with warfarin over a 12-week courseof treatment for DVT (after initial studies of enoxaparin withidraparinux demonstrated efficacy at all doses that was similar towarfarin). No clear dose–response relationship for efficacy wasshown with idraparinux, but a significant dose–response rela-tionship for major bleeding was shown (36). A large phase-IIItrial comparing the efficacy and safety of idraparinux with hep-arin or fondaparinux and dose-adjusted warfarin in both acuteand long-term treatment of DVT and PE has recently beencompleted (36).

3.1.2. SSR126517E(BiotinylatedIdraparinux)

This synthetic pentasaccharide being developed by Sanofi-Aventisexerts antithrombotic properties through AT-mediated inhibitionof FXa activity. It is identical to idraparinux with the exception ofa biotin moiety covalently affixed through a linker to the pen-tasaccharide structure so that anti-FXa activity can be neutral-ized in vivo by avidin. In vitro studies revealed that SSR126517Ebinds with high-affinity (Kd = 10.9 mol/l) to human AT andinhibits FXa in a concentration-dependent manner. It does notinhibit platelet aggregation or cross-react with antibodies fromsera of patients with HIT. In phase-I studies, the median time toreach maximum concentration was 4 h, with an absolute bioavail-ability of 100% and a half-life of approximately 200 h. Exposureof SSR126517E to avidin resulted in a rapid decrease of anti-Xaactivity and no serious adverse events (37).


3.2. Selective DirectFXa Inhibitors

Advantages of direct FXa inhibitors include the absence ofintermediary molecules such as AT that may potentially con-tribute to inconsistent anticoagulation, particularly in acute orinflammatory states. DX-9065a was the first of a class of small,synthesized, selective direct FXa inhibitors to undergo phase-IIclinical trials in arterial thrombosis (38, 39). Efforts to developorally available selective FXa inhibitors for VTE and preventionof stroke in AF patients are underway (see Table 5.1).

3.2.1. Razaxaban Razaxaban (BMS-561389), developed by Bristol-Myers Squibb(formerly Dupont), represents the first of a new class of syntheti-cally derived, small-molecule, oral-direct FXa inhibitors that donot require anticoagulant monitoring. In phase-I trials involv-ing young and elderly volunteers, it was well tolerated and wellabsorbed with only nuisance bleeding reported; FXa inhibitionand dose-dependent anticoagulation were noted. Razaxaban wasinvestigated for proof of principle in DVT prevention in patientsundergoing total knee replacement. In phase-IIb trials, razaxa-ban at doses of 25, 50, 75, or 100 mg twice daily administeredstarting 8 h after surgery was compared to enoxaparin at a doseof 30 mg twice daily initiated 12–24 h after surgery. The studyrevealed efficacy but an unacceptable risk/benefit profile at higherdoses (40). Razaxaban was discontinued for further developmentin March 2005 in lieu of an oral FXa inhibitor under develop-ment by the same company with a more favorable safety profile(see below).

3.2.2. Apixaban Apixaban (formerly BMS-562247 or DPC-AG0023) is an orallyactive, small-molecule direct FXa inhibitor being developed byBristol-Myers Squibb that has a more favorable safety profile thanrazaxaban. It is a highly potent inhibitor of human FXa, with a Kiof 0.08 ± 0.01 nM, and binds to serum proteins at a rate of 87%.It has a consistent oral absorption profile and linear pharmacoki-netics, with a maximal plasma concentration achieved within 3 hand an effective half-life of 9 h for twice-daily administration, and14 h for once-daily administration. It is eliminated via both hep-atic and renal routes. Apixaban had only modest effects on twotraditional markers of anticoagulation, international normalizedratio (INR) and activated partial thromboplastin time (aPTT).Phase-I clinical studies revealed mild bleeding and prolongedbleeding time, with no evidence of elevated transaminases (ala-nine transaminase or aspartate transaminase ≥ 5 times the upperlimit of the normal range). The safety profile of apixaban remainsto be determined in phase-II/III trials, and it is presently under-going phase-II clinical studies in elective total knee replacementsurgery patients and patients with acute DVT.

166 Mousa

The efficacy and safety of apixaban as a thromboprophylaxisin patients following total knee replacement was determined (41).Apixaban in doses of 2.5 mg twice daily or 5 mg once daily exhib-ited a promising benefit-risk profile as compared to current stan-dards of care following total knee replacement (41).

3.2.3. Rivaroxaban(Bayer HealthCare AGand J&J/Scios, Inc.)

Rivaroxaban (formerly BAY 59-7939) is a small-molecule, selec-tive oral-direct FXa inhibitor developed by Bayer for the pre-vention and treatment of thrombosis. Oral rivaroxaban may begiven in fixed once-daily doses, with potentially no coagula-tion monitoring. These properties, along with results from pre-clinical and clinical studies, suggest that the use of rivaroxabanmight be more advantageous than current treatments. Studiesin arterial and venous animal models have demonstrated thatrivaroxaban has potent antithrombotic properties without pro-longing bleeding times. In healthy subjects, rivaroxaban waswell tolerated with a predictable pharmacological profile anda low propensity for clinically relevant drug–drug interactions.In preclinical studies, endogenously generated FXa was inhib-ited with an IC50 of 21 ± 1 nM (42). This antithromboticeffect was demonstrated in different thrombosis models at dosesof 0.6–10 mg/kg, depending on the model and species, fol-lowing oral administration. In dogs, bioavailability ranged from60–86%.

In phase-I studies, rivaroxaban was rapidly absorbed [maxi-mum concentration (Cmax) achieved after 30 min] and well tol-erated (up to 80 mg single dose in healthy people) (43). Elimi-nation occurred with a terminal half-life of 4.86–9.15 h (steadystate). Prothrombin time (PT), aPTT, and HepTest were pro-longed in a dose-dependent manner, and there was no effect onbleeding time. In elderly men and women (>60 years), meanarea under the concentration–time curve and Cmax tended tobe about 20% higher. There were no drug–drug interactions orinduction of major cytochrome P450 (CYP) isoforms, with theexception of strong CYP3A4 inhibitors, and no prolongation ofQTc was observed.

Four large dose-ranging studies of rivaroxaban (ODIXa-HIP,ODIXa-HIP2, ODIXa-KNEE, and ODIXa-OD-HIP) have beencompleted, covering a 12-fold increase in dose, from 2.5 to 30 mgtwice daily and 5 to 40 mg once daily for VTE prevention follow-ing major orthopedic surgery (44, 45). The open-label phase-IIa ODIXa-HIP trial using mandatory venography confirmedproof of principle for rivaroxaban. Other studies have confirmedthe efficacy and safety of rivaroxaban as compared to enoxaparinunder double-blind, double-dummy conditions for VTE preven-tion in patients undergoing orthopedic surgery (44). Phase-IIstudies of rivaroxaban for the prevention of VTE after majororthopedic surgery support these findings and suggest that a total


daily dose range of 5–20 mg rivaroxaban has similar efficacy andsafety as enoxaparin, with 10 mg rivaroxaban once daily the opti-mal dose (46).

The drug development plan for rivaroxaban is aggressive,with a program of simultaneous investigation of multiple indica-tions rather than a sequential approach. At present, over 24,000patients (12,000 are predominantly post-operative orthopedicpatients) have been evaluated in completed phase-II and III trialsof rivaroxaban for thromboprophylaxis and treatment of DVT. Bythe time all the currently enrolling trials have concluded, morethan 50,000 patients will have been evaluated in randomizedcontrolled trials of rivaroxaban. The advantages of rivaroxabaninclude the potential for once-daily dosing for all indications, norequired dose adjustment for body weight, no known interac-tions with common cardiovascular medications, a relatively safepharmacodynamic profile with respect to bleeding risk and hep-atotoxicity, no clinically significant interaction with aspirin, andthe ability to bridge with LMWH when necessary. On the otherhand, rivaroxaban is partially renally cleared and will require doseadjustment in those with grade-III chronic kidney disease andis not being studied in patients with a creatinine clearance lessthan 30 ml/min. In addition, since rivaroxaban metabolism isaffected by potent cytochrome P450 3A4 inhibitors such as keto-conazole and clarithromycin, and protease inhibitors, use will berestricted in certain special populations. Nonetheless, after exten-sive phase-II, and now emerging phase-III trial data, it appearsthat rivaroxaban is effective in preventing and treating VTE witha bleeding risk comparable to other anticoagulants. The resultsof randomized trials evaluating rivaroxaban for the prevention ofstroke and non-central nervous system (CNS) embolism in AFand secondary prevention of acute coronary syndromes are cur-rently ongoing.

While there are several other oral anticoagulants in devel-opment, none have been evaluated as extensively or in as manypatients as rivaroxaban. Rivaroxaban has the advantage that it canbe dosed once a day, which has been shown to improve patientcompliance and outcomes (47, 48). Despite once-daily dosing,rivaroxaban has a half-life that is considerably shorter than otheroral FXa inhibitors, which is advantageous in the event of bleed-ing or an urgent need to discontinue anticoagulation. Unlike theDTI dabigatran, the bioavailability of rivaroxaban is very good,and there is low risk of drug–drug interactions, including withmedications that alter gastric pH, which are taken chronically by3% of the US population (49). Perhaps more importantly, rivarox-aban has been shown to have no effect on platelet aggregation,and its pharmacokinetic profile is unaffected by aspirin and othernotable cardiovascular medications such as digoxin. Finally, afterclinical investigation in thousands of patients, rivaroxaban appears

168 Mousa

to have no significant hepatotoxicity and a bleeding risk compa-rable to other conventional anticoagulants.

Rivaroxaban could potentially be used in several clinical indi-cations and disease states, including VTE prophylaxis, long-termtreatment of DVT, PE, and the prevention of stroke and non-CNS embolism in patients with AF and possibly acute coronarysyndromes. In order to understand the efficacy and safety of atherapeutic across such a wide range of patient populations, sev-eral large simultaneous studies evaluating rivaroxaban in bothvenous and arterial thromboembolism are under way. The great-est therapeutic impact of rivaroxaban might be in providing amuch-needed and attractive alternative to warfarin. While formalcost-effective analyses are not yet available, avoiding the intense,costly, and frequent monitoring required with VKAs, as well asthe potential to reduce adverse vascular events precipitated by thenarrow therapeutic window of VKAs, will most likely translateinto a significant improvement in quality of life and cost savings.While further data (especially large phase-III trials) and cautionare required, there is reason for optimism. Rivaroxaban may verywell be the long-awaited alternative to warfarin.

3.2.4. YM-150 Astellas (formerly Yamanouchi) has developed YM-150, an oral-selective FXa inhibitor for DVT prevention. The compoundhas an immediate antithrombotic effect after oral administra-tion, with a dose-dependent response and prolongation of PT;no significant food interactions have been noted. In a phase-II dose-escalation study in patients undergoing elective primaryhip replacement surgery, YM-150 (3, 10, 30, or 60 mg POonce daily) given 6–10 h after surgery for 7–10 days was com-pared to enoxaparin administered at a dose of 40 mg SQ oncedaily 12 h before surgery (50). There was no major bleed-ing, and the median incidence of VTE ranged from 52% in the3-mg group to 19% in the 60-mg group. Overall, the drugappears to be safe and well tolerated. A dose-escalation studyof YM150 in the prevention of VTE in elective primary hipreplacement surgery was also carried out (51). YM150, admin-istered at doses of 10–60 mg daily starting 6–10 h after pri-mary hip replacement was shown to be safe, well tolerated, andeffective.

3.2.5. DU-176b Daiichi Sankyo is developing DU-176b, an oral FXa inhibitor forthe treatment of thrombotic disorders. Preclinical data in mousemodels revealed potent antithrombotic effects in AT-positive andAT-deficient mice (52). In rat models, DU-176b at doses of 0.05–1.25 mg/kg/h prevented arterial and venous thrombosis.

In terms of clinical studies, DU-1766 in a single 60-mg dosewas given to healthy males. The drug inhibited FXa activity,reduced thrombin generation, prolonged PT, aPTT and INR, and


reduced venous and arterial thrombosis by 28 and 26%, respec-tively, in a Badimon chamber. Further studies are planned.

3.2.6. LY-517717 LY-517717, an indol-6-yl-carbonyl derivative, is the lead in aseries of oral-selective FXa inhibitors being developed by Eli Lillyas part of a research collaboration with Protherics for the poten-tial treatment of thromboembolic diseases. It is 1000-fold moreselective as an FXa inhibitor than other serine proteases, with a Kiof 5 nM. The oral bioavailability of LY-517717 is approximately25–82%, with a plasma half-life of 7–10 h. In a rat atrioventricu-lar shunt model, the median effective dose was 5–10 mg/kg PO,and absorption in dogs indicated no bleeding issues. In a phase-Istudy, LY-517717 was found to be well tolerated and suitable foronce-daily administration. In a dose-escalating study, 511 patientsundergoing hip or knee replacement surgery were randomized toreceive one of the six oral doses of LY-517717 (25, 50, 75, 100,125, or 150 mg), or enoxaparin (40 mg SQ daily), started preop-eratively, for 6–10 doses. The 100, 125, and 150-mg dose groupswere non-inferior to enoxaparin in the incidence of symptomaticor venographically proven DVT or PE (53). The compound pro-duced a dose-dependent prolongation of PT and was well toler-ated, and there were no differences in bleeding risk as comparedto enoxaparin (53).

3.3. Selective, DirectFIXa and FXIInhibitors

Although the development of direct FIXa inhibitors is at an earlierphase than direct FXa inhibitors, the theoretical advantages aresimilar. TTP 889, manufactured by Trans Tech Pharma, is an oral,direct FIXa inhibitor with a half-life of 20 h, enabling once-dailydosing. A phase-II proof of principle study for VTE prevention inhip fracture surgery, the FIXIT trial, recently completed enroll-ment of 206 patients who received standard in-hospital throm-boprophylaxis. Efficacy and safety are being compared betweenpatients randomized to receive TTP 889 versus placebo for up to3 weeks post-discharge (54).

Development of FXIa inhibitors is currently at the preclinicallevel. BMS-262084 is an irreversible and selective small-moleculeinhibitor of FXIa, with an IC50 of 2.8 nM against human FXIa.The effect of inhibiting activated blood coagulation FXIa withBMS-262084 has been determined in rat models of thrombosisand hemostasis (55). BMS-262084 doubled aPTT in human andrat plasma at concentrations of 0.14 and 2.2 μM, respectively.Consistent with FXIa inhibition, PT was unaffected at concentra-tions up to 100 μM. BMS-262084 administered by intravenousloading with sustained infusion was effective against FeCl(2)-induced thrombosis in both the vena cava and the carotid artery.In contrast, doses of up to 24 mg/kg +24 mg/kg/h had noeffect on TF-induced venous thrombosis or ex vivo PT. Dosesof up to 24 mg/kg+24 mg/kg/h did not significantly prolong

170 Mousa

bleeding time provoked by puncture of small mesenteric bloodvessels, template incision of the renal cortex, or cuticle incision.These results demonstrate that pharmacologic inhibition of FXIaachieves antithrombotic efficacy with minimal effects on pro-voked bleeding (55).

4. Inhibitors ofThrombin Activity

Thrombin is the central serine protease in hemostasis. The mech-anisms of action of thrombin involve coagulation, platelet activa-tion, fibrinolysis, and vascular cell biology. In addition to its majorrole in fibrin formation and activation of FXIII, which cross-linksfibrin, thrombin is essential for feedback activation of other coag-ulation factors such as FV, FVIII, and FIX (56). Thrombin isinvolved in platelet activation and subsequent aggregation (53)and can act as an anticoagulant by binding to thrombomodulin,which converts protein C to its active form, inactivating FVaand FVIIIa. Thrombomodulin-bound thrombin regulates coag-ulation through activation of thrombin-activatable fibrinolysisinhibitor (TAFI) and subsequent down-regulation of fibrinolysis.Predictions that thrombin inhibitors would be more effective thanFXa inhibitors in arterial thromboembolic disease (where throm-bin has a key role in platelet activation) and less effective in VTEhave not been borne out by clinical data. Given its central role inthe coagulation cascade, inhibitors of thrombin activity – whethermediated by AT or acting directly on the active site – represent animportant class of anticoagulant drugs in our armamentarium.

4.1. IndirectThrombin Inhibitors

SNAC/Heparin. SNAC (sodium N-[8(2-hydroxybenzoyl)amino]caprylate), developed by Emisphere Technologies, enablesmacromolecule delivery of the large negatively charged andpoorly absorbed heparin molecule through a noncovalent com-plex with heparin, allowing passive, transcellular absorption.SNAC itself has no pharmacological activity. Phase-I studiesrevealed that in doses of up to 10.5 g SNAC/150,000 U hep-arin, the compound is well tolerated, with nausea being the onlysignificant adverse event observed. Dose-dependent increases inaPTT and anti-FXa levels were also observed, suggesting thatboth AT-mediated thrombin and FXa inhibition play a role in theanticoagulant effects of the drug (57). In addition, there was anapparent food and diurnal effect, but no age effect.

The PROTECT study was a large phase-III study in 2,264 hipreplacement patients with two SNAC treatment arms (low-doseSNAC and high-dose SNAC) for 30 days versus 10 days ofenoxaparin at a dose of 30 mg SC every 12 h (58). Mandatory


venography on days 27–30 revealed an overall VTE rate of31.8% in the low-dose SNAC group, 29.7% in the high-dosegroup, and 26.1% in the enoxaparin group. The rates of prox-imal DVT/PE were 18.6% in the low-dose group and 13.8%in the high-dose group, both of which were significantly higherthan in the enoxaparin group (12.7%, P = 0.013 and P = 0.045,respectively). There was overall poor compliance in 22.1% of thepatients on the low-dose regimen and in 31.4% of the patientson the high-dose regimen, suggesting that substance compli-ance was a key factor in failure to achieve proof of principlefor the use of SNAC/heparin in VTE prevention as an indi-rect FIIa/Xa inhibitor. Improved formulations of heparin in soliddosage form are currently under clinical investigation (59, 60).An orally administrable chemical conjugate of heparin andhydrophobic deoxycholic acid, referred to as LHD, has also beendeveloped. LHD was pre-formulated with dimethyl sulfoxide assolubilizer to further improve oral bioavailability (9.1% in mon-keys). LHD was absorbed mainly in the jejunum and ileum ofthe small intestine, although it is in the ileum that absorptionwas most notable. Mechanistic studies of LHD absorption usingCaco-2 cell monolayers, which mimic the intestine, demonstratedthat the high permeability of LHD is mediated by passive diffu-sion through the transcellular route, and permeation is affected inpart by bile acid transporters. These results demonstrated the fea-sibility of using chemically modified heparin for long-term oraladministration as an effective therapy for VTE (59). A morerecent clinical study determined the true pharmacokinetics forinjectable versus oral heparin (60).

4.2. Direct ThrombinInhibitors

The development of DTIs was driven by three major factors:increasing recognition of immune thrombocytopenia as a poten-tially severe complication of heparin use (61), the notion thatheparin-AT inhibition of thrombin produces only weak inhibitionof cell surface- or clot-bound thrombin, which is active and canbe released during fibrinolysis (62), and the non-specific bindingproperties of heparin, necessitating frequent monitoring. Hence,non-AT-based thrombin inhibitors with improved safety profilesover heparin, the ability to inhibit surface- or clot-bound throm-bin, and predictable dose responses would be advantageous inthe clinical setting. Furthermore, oral formulations of these drugswould be a major advantage. DTIs could be ideal drugs for thetreatment of HIT, as this condition is characterized by the gener-ation of large amounts of thrombin.

A theoretical concern about DTIs is that they could inhibitthe anticoagulant properties of thrombin, namely inhibitionof the thrombin–thrombomodulin-mediated negative feedbackmechanism of the protein C system, with the possibility ofrebound hypercoagulability (63).

172 Mousa

Four parenteral DTIs have emerged: lepirudin, bivalirudin,argatroban, and melagatran, the first three of which have beenapproved for clinical use. Lepirudin is a naturally occurring biva-lent DTI indicated for thromboprophylaxis of HIT. Argatrobanis a prototype noncovalent, reversible, small-molecule DTI indi-cated for thromboprophylaxis or treatment of HIT. Melagatran isthe active form of the oral, prodrug, small-molecule DTI xime-lagatran, discussed below. All of these drugs have limitations interms of parenteral use, limited indications, need for frequentmonitoring, and high cost.

4.3. SelectiveOral-Direct ThrombinInhibitors

4.3.1. Ximelagatran Ximelagatran, developed by AstraZeneca, represents the first of anew class of orally active, small-molecule DTIs to reach late-stagedevelopment with limited clinical indications for VTE preven-tion. Ximelagatran is a hydrophilic prodrug that is converted by acytochrome-P450-independent liver enzyme system to its activeform melagatran. Bioavailability of ximelagatran is approximately20%, and the half-life is 4–5 h in patients. It can be administeredtwice daily and does not require anticoagulant monitoring or doseadjustment. Ximelagatran was studied extensively in a large phase-III trial for VTE prevention and treatment and was found tobe either superior or equivalent to warfarin in terms of efficacy(64–69). However, initial long-term data with ximelagatranrevealed elevated liver enzymes (approximately 6%). Based uponthis and other considerations, it was not approved by the USFDA. It was, however, approved in other countries for short-term, post-orthopedic thromboprophylaxis. In February 2006,AstraZeneca withdrew ximelagatran from the world market dueto continuing concerns about severe liver toxicity associated withlong-term use.

4.3.2. Dabigatran Dabigatran etexilate is a small-molecule, orally active, prodrugDTI developed by Boehringer Ingelheim that has reached late-stage clinical development. It is rapidly absorbed and convertedto the active form, dabigatran. It has linear characteristics in termsof concentration and global coagulation parameters, includingthrombin clotting time, INR, and ecarin clotting time. Dabiga-tran has a Ki of 4.5 ± 0.2 nmol/l, peak plasma concentration 2 hpost-dose, and a half-life of approximately 14–17 h after multipledose administration (70). It is metabolized mainly (80–85%) byrenal excretion.

The BISTRO II study was a multicenter, parallel group,double-blind, dose-finding study for VTE prevention in 1,949patients undergoing total hip or knee replacement (71). Patientswere randomized to receive dabigatran (50 mg, 150 mg, or


225 mg twice daily, or 300 mg once daily) starting 14 h aftersurgery. The comparator was enoxaparin (40 mg once daily) initi-ated 12 h prior to surgery. A significant dose-dependent decreasein DVT was observed with increasing doses of dabigatran (P <0.001). Compared to enoxaparin, DVT was significantly lower inpatients receiving dabigatran 150 mg twice daily [odds ratio (OR)0.47, P = 0.0007], 300 mg once daily (OR 0.61, P = 0.02), and225 mg twice daily (OR 0.47, P = 0.0007). Major bleeding waslower with the low dose of dabigatran (0.3% vs. 2.0%, P = 0.047),but elevated at higher doses, with trends almost reaching statis-tical significance in those receiving the 300 mg dabigatran dose(4.7%, P = 0.051). In terms of adverse events, the incidenceof elevated alanine aminotransferase (>3 times ULN) was lowerin the dabigatran groups (1.5–3.1%) than in the enoxaparingroup (7.4%). There were no cases of clinically relevant throm-bocytopenia. The authors concluded that dabigatran started inthe early post-operative period was effective and safe across awide range of doses. In addition, the frequency and extent ofsevere hepatic abnormalities were lower than those observedwith ximelagatran. Dabigatran is currently undergoing exten-sive phase-III evaluation for VTE prevention, treatment, andsecondary thromboprophylaxis through the RE-VOLUTIONprogram.

A randomized, double-blind, non-inferiority trial was con-ducted comparing dabigatran etexilate to enoxaparin for preven-tion of VTE after total hip replacement (72). Patients (3,494total) undergoing total hip replacement were randomized intotreatment for 28–35 days with dabigatran etexilate 220 mg(n=1157) or 150 mg (n=1174) once daily, starting with a half-dose 1–4 h after surgery, or SC enoxaparin (40 mg) once daily(n=1162), starting the evening before surgery. The primary effi-cacy outcome was the composite of total VTE (venographic orsymptomatic) and death from all causes during treatment. Onthe basis of the absolute difference in rates of VTE with enoxa-parin versus placebo, the non-inferiority margin for the differ-ence in rates of thromboembolism was defined as 7.7%. Bothdoses of dabigatran were non-inferior as compared to enoxaparin.There was no significant difference in major bleeding rates witheither dose of dabigatran etexilate as compared with enoxaparin(72). The frequency of increased liver enzyme concentrations andacute coronary events during the study did not differ significantlybetween the groups. The study concluded that oral dabigatranetexilate was as effective as enoxaparin in reducing the risk ofVTE after total hip replacement surgery, with a similar safetyprofile (72).

4.3.3. TGN-167 TGN-167 (TRI-50c-04) is an oral thrombin inhibitor beingdeveloped by Trigen Holdings for the potential treatment of

174 Mousa

thrombosis. A controlled release formulation of the drug is alsobeing developed with Eurand for long-term treatment of throm-bosis. The compound produces a marked increase in thrombinclotting time, with minimal effects on aPTT. A double-blind,phase-I, dose-escalation study with 20 volunteers showed thedrug to be well tolerated, with no significant adverse eventsreported (73). At 600 mg, all dosed subjects achieved effectiveanticoagulant activity in vitro. Trigen is planning to continueTGN-167 into phase-II studies.

5. Conclusions

The antithrombotic management of VTE will undergo significantchanges in the next 5-10 years. Limitations of existing parenteraland oral anticoagulants has led to the development of new agentsdesigned to target specific procoagulant complexes in the coag-ulation pathway, inhibiting coagulation initiation, coagulationpropagation, or thrombin activity. With respect to efficacy, dur-ing acute treatment of VTE, newer antithrombotic agents mustexhibit at least non-inferiority in a methodologically sound studyas compared to the existing parenteral agent of choice, LMWH,and the emerging agent fondaparinux. This is true particularly inhigh-risk venous thromboses, such as ileofemoral VTE, PE, orVTE associated with cancer. For long-term VTE treatment, thereis a need to improve upon existing oral anticoagulants, namely,VKAs. Target-selective oral agents must exhibit an improvedsafety profile (especially as it pertains to major or clinically signifi-cant bleeding), ease of use, and tolerability as compared to VKAs.If successful, emerging oral anticoagulants could negate the tra-ditional distinction of acute versus long-term treatment of VTE,as they could potentially be used throughout the spectrum of dis-ease without the need for overlap with parenteral therapies (52).Lastly, any new long-term anticoagulant must be safely toleratedin combination with antiplatelet agents, as an increasingly agingpopulation will be prone to arterial as well as venous thromboem-bolic disease. Cost considerations are also important, especiallyfrom a populational perspective.

Newer agents should, in theory, fulfill the following require-ments of an ideal anticoagulant: a rapid onset with predictableresponse characteristics, predictable pharmacokinetics, pharmaco-dynamics with low plasma protein binding, no required moni-toring, a half-life that provides both safety and ease of use (par-ticularly during temporary withdrawal), lack of food or druginteractions, an excellent safety profile (particularly with respectto immune-mediated thrombocytopenia, hepatotoxicity, and


potential for thrombotic rebound phenomenon), and reversibil-ity or availability of an antidote. In addition, oral agents withpredictable intestinal absorption/bioavailability used in a simple,fixed-dose once or twice daily regimen and for which compli-ance can be monitored would be even more advantageous. Atthis time, drugs at the most advanced stage of development withrespect to VTE management include the parenteral indirect FXainhibitor idraparinux and biotinylated idraparinux, the oral DTIdabigatran, and the oral selective direct FXa inhibitors rivarox-aban and apixaban. Whether there are inherent advantages inblocking initial thrombin formation via the prothrombinase com-plex early in the coagulation system or blocking thrombin directlyand preventing feedback amplification is still a matter of debate,as is the notion of whether there is any clinically meaningfuleffect of small-molecule DTIs that target both clot-bound andfree thrombin. Long-term clinical data with respect to efficacy ofanti-Xa inhibitors will be available shortly, while long-term dataare currently available on the efficacy of direct thrombin inhi-bition. The lessons from ximelagatran reveal the importance oflong-term safety data in different patient populations. Ximelaga-tran had shown significant potential as a possible replacement towarfarin therapy, but was withdrawn because of potential livertoxicity. In contrast, dabigatran appears to have a better safetyprofile and has recently entered a phase-III randomized clinicaltrial for AF. Oral direct FXa inhibitors (rivaroxaban, apixaban, andothers) may prove to be more potent and safe. Selective inhibitorsof specific coagulation factors involved in the initiation and prop-agation of the coagulation cascade (FIXa, FVIIa, circulating TF)are at an early stage of development. Additional new agents inclinical development include NAPc2, protein C derivatives, andsoluble thrombomodulin.

References

1. Anderson, F.A., Jr., and Spencer, F.A. (2003)Risk factors for venous thromboembolismCirculation 107, I9–16.

2. Kearon, C. (2003) Natural history of venousthromboembolism Circulation 107, I22–30.

3. Warkentin, T.E., Levine, M.N., Hirsh, J.,Horsewood, P., Roberts, R.S., Gent, M., andKelton, J.G. (1995) Heparin-induced throm-bocytopenia in patients treated with low-molecular-weight heparin or unfractionatedheparin N Engl J Med 332, 1330–5.

4. Spyropoulos, A.C. (2006) Managing oralanticoagulation requires expert experienceand clinical evidence J Thromb Thromboly 21,91–4.

5. Bates, S.M., and Ginsberg, J.S. (1997) Anti-coagulants in pregnancy: fetal effects Bail-lieres Clin Obstet Gynaecol 11, 479–88.

6. Hutten, B.A., Prins, M.H., Gent, M., Gins-berg, J., Tijssen, J.G., and Buller, H.R.(2000) Incidence of recurrent thromboem-bolic and bleeding complications amongpatients with venous thromboembolism inrelation to both malignancy and achievedinternational normalized ratio: a retrospec-tive analysis J Clin Oncol 18, 3078–83.

7. Palareti, G., Legnani, C., Lee, A., Manotti,C., Hirsh, J., D’Angelo, A., Pengo, V., Moia,M., and Coccheri, S. (2000) A comparison ofthe safety and efficacy of oral anticoagulation

176 Mousa

for the treatment of venous thromboembolicdisease in patients with or without malig-nancy Thromb Haemost 84, 805–10.

8. Bates, S., Greer, I., Hirsch, J., and Gins-berg, J. (2004) Use of antithrombotic agentsduring pregnancy. The seventh ACCP con-ference on antithrombotic and thrombolytictherapy Chest 126, 627S–44S.

9. Greer, I.A., and Nelson-Piercy, C. (2005)Low-molecular-weight heparins for throm-boprophylaxis and treatment of venousthromboembolism in pregnancy: a system-atic review of safety and efficacy Blood 106,401–7.

10. Lee, A.Y., Levine, M.N., Baker, R.I., Bow-den, C., Kakkar, A.K., Prins, M., Rickles,F.R., Julian, J.A., Haley, S., Kovacs, M.J.,and Gent, M. (2003) Low-molecular-weightheparin versus a coumarin for the preven-tion of recurrent venous thromboembolismin patients with cancer N Engl J Med 349,146–53.

11. Weitz, J.I., Hirsh, J., and Samama, M.(2004) New anticoagulant drugs. The sev-enth ACCP conference on antithromboticand thrombolytic therapy Chest 126, 265S–86S.

12. Dahl, O.E. (1999) Mechanisms of hyperco-agulability Thromb Haemost 82, 902–6.

13. Fernandez, P.M., Patierno, S.R., and Rickles,F.R. (2004) Tissue factor and fibrin in tumorangiogenesis Semin Thromb Hemost 30,31–44.

14. Girard, T.J., and Nicholson, N.S. (2001) Therole of tissue factor/factor VIIa in the patho-physiology of acute thrombotic formationCurr Opin Pharmacol 1, 159–63.

15. Hirsh, J., O’Donnell, M., and Weitz, J.I.(2005) New anticoagulants Blood 105,453–63.

16. Weitz, J.I., and Bates, S.M. (2005) New anti-coagulants J Thromb Haemost 3, 1843–53.

17. Mousa, S.A., and Mohamed, S. (2004) Inhi-bition of endothelial cell tube formation bythe low molecular weight heparin, tinzaparin,is mediated by tissue factor pathway inhibitorThromb Haemost 92, 627–33.

18. Lee, A., Agnelli, G., Buller, H., Ginsberg,J., Heit, J., Rote, W., Vlasuk, G., Costan-tini, L., Julian, J., Comp, P., van Der Meer,J., Piovella, F., Raskob, G., and Gent, M.(2001) Dose-response study of recombinantfactor VIIa/tissue factor inhibitor recombi-nant nematode anticoagulant protein c2 inprevention of postoperative venous throm-boembolism in patients undergoing totalknee replacement Circulation 104, 74–8.

19. Giugliano, R.P., Wiviott, S.D., Stone, P.H.,Simon, D.I., Schweiger, M.J., Bouchard,

A., Leesar, M.A., Goulder, M.A., Deitcher,S.R., McCabe, C.H., and Braunwald, E.(2007) Recombinant nematode anticoagu-lant protein c2 in patients with non-ST-segment elevation acute coronary syndrome:the ANTHEM-TIMI-32 trial J Am Coll Car-diol 49, 2398–407.

20. Moons, A.H., Peters, R.J., Bijsterveld, N.R.,Piek, J.J., Prins, M.H., Vlasuk, G.P., Rote,W.E., and Buller, H.R. (2003) Recombi-nant nematode anticoagulant protein c2, aninhibitor of the tissue factor/factor VIIacomplex, in patients undergoing electivecoronary angioplasty J Am Coll Cardiol 41,2147–53.

21. Edgington, T.S., Mackman, N., Brand, K.,and Ruf, W. (1991) The structural biologyof expression and function of tissue factorThromb Haemost 66, 67–79.

22. Arnold, C.S., Parker, C., Upshaw, R., Prydz,H., Chand, P., Kotian, P., Bantia, S., andBabu, Y.S. (2006) The antithrombotic andanti-inflammatory effects of BCX-3607, asmall molecule tissue factor/factor VIIainhibitor Thromb Res 117, 343–9.

23. Lazarus, R.A., Olivero, A.G., Eigenbrot, C.,and Kirchhofer, D. (2004) Inhibitors of tis-sue factor. Factor VIIa for anticoagulant ther-apy Curr Med Chem 11, 2275–90.

24. Buckman, B.O., Chou, Y.L., McCarrick, M.,Liang, A., Lentz, D., Mohan, R., Mor-rissey, M.M., Shaw, K.J., Trinh, L., andLight, D.R. (2005) Solid-phase synthesis ofnaphthylamidines as factor VIIa/tissue fac-tor inhibitors Bioorg Med Chem Lett 15,2249–52.

25. Suleymanov, O.D., Szalony, J.A., Saly-ers, A.K., LaChance, R.M., Parlow, J.J.,South, M.S., Wood, R.S., and Nicholson,N.S. (2003) Pharmacological interruptionof acute thrombus formation with mini-mal hemorrhagic complications by a smallmolecule tissue factor/factor VIIa inhibitor:comparison to factor Xa and thrombin inhi-bition in a nonhuman primate thrombosismodel J Pharmacol Exp Ther 306, 1115–21.

26. Bajaj, M.S., and Bajaj, S.P. (1997) Tissue fac-tor pathway inhibitor: potential therapeuticapplications Thromb Haemost 78, 471–7.

27. Mousa, S.A., Fareed, J., Iqbal, O., andKaiser, B. (2004) Tissue factor pathwayinhibitor in thrombosis and beyond MethodsMol Med 93, 133–55.

28. Sandset, P.M., Abildgaard, U., and Larsen,M.L. (1988) Heparin induces release ofextrinsic coagulation pathway inhibitor (EPI)Thromb Res 50, 803–13.

29. Amirkhosravi, A., Meyer, T., Amaya, M.,Davila, M., Mousa, S.A., Robson, T., and


Francis, J.L. (2007) The role of tissuefactor pathway inhibitor in tumor growthand metastasis Semin Thromb Hemost 33,643–52.

30. Mousa, S.A., Linhardt, R., Francis, J.L.,and Amirkhosravi, A. (2006) Anti-metastaticeffect of a non-anticoagulant low-molecular-weight heparin versus the standard low-molecular-weight heparin, enoxaparinThromb Haemost 96, 816–21.

31. Sierko, E., Wojtukiewicz, M.Z., and Kisiel,W. (2007) The role of tissue factor pathwayinhibitor-2 in cancer biology Semin ThrombHemost 33, 653–9.

32. Biemond, B.J., Friederich, P.W., Levi, M.,Vlasuk, G.P., Buller, H.R., and ten Cate, J.W.(1996) Comparison of sustained antithrom-botic effects of inhibitors of thrombin andfactor Xa in experimental thrombosis Circu-lation 93, 153–60.

33. Buller, H.R., Davidson, B.L., Decousus, H.,Gallus, A., Gent, M., Piovella, F., Prins,M.H., Raskob, G., Segers, A.E., Cariou,R., Leeuwenkamp, O., and Lensing, A.W.(2004) Fondaparinux or enoxaparin for theinitial treatment of symptomatic deep venousthrombosis: a randomized trial Ann InternMed 140, 867–73.

34. Buller, H.R., Davidson, B.L., Decousus, H.,Gallus, A., Gent, M., Piovella, F., Prins,M.H., Raskob, G., van den Berg-Segers,A.E., Cariou, R., Leeuwenkamp, O., andLensing, A.W. (2003) Subcutaneous fon-daparinux versus intravenous unfractionatedheparin in the initial treatment of pulmonaryembolism N Engl J Med 349, 1695–702.

35. Herbert, J.M., Herault, J.P., Bernat, A.,van Amsterdam, R.G., Lormeau, J.C., Peti-tou, M., van Boeckel, C., Hoffmann, P.,and Meuleman, D.G. (1998) Biochemicaland pharmacological properties of SANORG34006, a potent and long-acting syntheticpentasaccharide Blood 91, 4197–205.

36. Reiter, M., Bucek, R.A., Koca, N., Heger,J., and Minar, E. (2003) Idraparinux andliver enzymes: observations from the PER-SIST trial Blood Coagul Fibrinolysis 14, 61–5.

37. Buller, H.R., Cohen, A.T., Davidson, B.,Decousus, H., Gallus, A.S., Gent, M., Pillion,G., Piovella, F., Prins, M.H., and Raskob,G.E. (2007) Idraparinux versus standardtherapy for venous thromboembolic diseaseN Engl J Med 357, 1094–104.

38. Alexander, J.H., Dyke, C.K., Yang, H.,Becker, R.C., Hasselblad, V., Zillman, L.A.,Kleiman, N.S., Hochman, J.S., Berger, P.B.,Cohen, E.A., Lincoff, A.M., Saint-Jacques,H., Chetcuti, S., Burton, J.R., Buergler,J.M., Spence, F.P., Shimoto, Y., Robertson,

T.L., Kunitada, S., Bovill, E.G., Armstrong,P.W., and Harrington, R.A. (2004) Ini-tial experience with factor-Xa inhibitionin percutaneous coronary intervention: theXaNADU-PCI Pilot J Thromb Haemost 2,234–41.

39. Alexander, J.H., Yang, H., Becker, R.C.,Kodama, K., Goodman, S., Dyke, C.K.,Kleiman, N.S., Hochman, J.S., Berger, P.B.,Cohen, E.A., Lincoff, A.M., Burton, J.R.,Bovill, E.G., Kawai, C., Armstrong, P.W.,and Harrington, R.A. (2005) First experi-ence with direct, selective factor Xa inhibitionin patients with non-ST-elevation acute coro-nary syndromes: results of the XaNADU-ACS trial J Thromb Haemost 3, 439–47.

40. Lassen, M., Davidson, B., Gallus, A., Pineo,G., Ansell, J., and Deitchman, D.B.a.A.(2003) A phase II randomized, double-blind, five-arm, parallel group, dose–responsestudy of a new oral directly-acting factor Xainhibitor, razaxaban, foe the prevention ofdeep vein thrombosis in knee replacementsurgery Blood 102, Abstract 41.

41. Lassen, M.R., Davidson, B.L., Gallus, A.,Pineo, G., Ansell, J., and Deitchman, D.(2007) The efficacy and safety of apixaban,an oral, direct factor Xa inhibitor, as throm-boprophylaxis in patients following total kneereplacement J Thromb Haemost 5, 2368–75.

42. Perzborn, E., Strassburger, J., Wilmen, A.,Pohlmann, J., Roehrig, S., Schlemmer, K.H.,and Straub, A. (2005) In vitro and in vivostudies of the novel antithrombotic agentBAY 59-7939–an oral, direct Factor Xainhibitor J Thromb Haemost 3, 514–21.

43. Kubitza, D., Becka, M., Voith, B., Zuehls-dorf, M., and Wensing, G. (2005) Safety,pharmacodynamics, and pharmacokinetics ofsingle doses of BAY 59-7939, an oral, directfactor Xa inhibitor Clin Pharmacol Ther 78,412–21.

44. Eriksson, B.I., Borris, L., Dahl, O.E., Haas,S., Huisman, M.V., Kakkar, A.K., Misselwitz,F., and Kalebo, P. (2006) Oral, direct Fac-tor Xa inhibition with BAY 59-7939 for theprevention of venous thromboembolism aftertotal hip replacement J Thromb Haemost 4,121–8.

45. Turpie, A.G., Fisher, W.D., Bauer, K.A.,Kwong, L.M., Irwin, M.W., Kalebo, P., Mis-selwitz, F., and Gent, M. (2005) BAY 59-7939: an oral, direct factor Xa inhibitor forthe prevention of venous thromboembolismin patients after total knee replacement.A phase II dose-ranging study J ThrombHaemost 3, 2479–86.

46. Laux, V., Perzborn, E., Kubitza, D., andMisselwitz, F. (2007) Preclinical and clinical

178 Mousa

characteristics of rivaroxaban: a novel, oral,direct factor Xa inhibitor Semin ThrombHemost 33, 515–23.

47. Claxton, A.J., Cramer, J., and Pierce, C.(2001) A systematic review of the associa-tions between dose regimens and medicationcompliance Clin Ther 23, 1296–310.

48. Richter, A., Anton, S.E., Koch, P., and Den-nett, S.L. (2003) The impact of reducingdose frequency on health outcomes Clin Ther25, 2307–35, Discussion 6.

49. Jacobson, B.C., Ferris, T.G., Shea, T.L.,Mahlis, E.M., Lee, T.H., and Wang, T.C.(2003) Who is using chronic acid suppres-sion therapy and why? Am J Gastroenterol 98,51–8.

50. Eriksson, B.I., Turpie, A.G., Lassen, M.R.,Prins, M.H., Agnelli, G., Kalebo, P., Gaillard,M.L., and Meems, L. (2007) A dose escala-tion study of YM150, an oral direct factor Xainhibitor, in the prevention of venous throm-boembolism in elective primary hip replace-ment surgery J Thromb Haemost 5, 1660–5.

51. Eriksson, B.I., Turpie, A., Lassen, M., Prins,M., Agnelli, G., Gaillard, M., and Meems,B. (2005) YM150, an oral direct factor Xainhibitor, as prophylaxis for venous throm-boembolism in patients with elective primaryhip replacement surgery. A dose escalationstudy Bood 106, 1865.

52. Fukuda, F., Honda, Y., Matsumoto, C.,Sugiyama, N., Matsushita, T., Yanada, M.,Morishima, Y., and Shibano, T. (2005)Impact of antithrombin deficiency on effi-ciencies of DU-176b, a novel orally activedirect factor Xa inhibitor, and antithrombindependent anticoagulants, fondaparinux, andheparin Blood 106, 1874.

53. Agnelli, G., Haas, S., Krueger, K., Bedding,A., and Brandt, J. (2005) A phase II studyof the safety and efficacy of a novel oral fXainhibitor (LY517717) for the prevention ofvenous thromboembolism following TKR orTHR Blood 106, 85a.

54. Rothlein, R., Shen, J., Naser, N.,Gohimukkula, D., Caligan, T., Andrews, R.,Schmidt, A., Rose, E., and Mjalli, A. (2005)TTP889, a novel orally active partial inhibitorof FIXa inhibits clotting in two A/V shuntmodels without prolonged bleeding Blood106, 1886.

55. Schumacher, W.A., Seiler, S.E., Steinbacher,T.E., Stewart, A.B., Bostwick, J.S., Hartl,K.S., Liu, E.C., and Ogletree, M.L. (2007)Antithrombotic and hemostatic effects of asmall molecule factor XIa inhibitor in ratsEur J Pharmacol 570, 167–74.

56. Lundblad, R.L., Bradshaw, R.A., Gabriel, D.,Ortel, T.L., Lawson, J., and Mann, K.G.

(2004) A review of the therapeutic uses ofthrombin Thromb Haemost 91, 851–60.

57. Baughman, R.A., Kapoor, S.C., Agarwal,R.K., Kisicki, J., Catella-Lawson, F., andFitzGerald, G.A. (1998) Oral delivery ofanticoagulant doses of heparin. A ran-domized, double-blind, controlled study inhumans Circulation 98, 1610–5.

58. Hull, D., Kakkar, A., Marder, V., Pineo, G.F.,Goldberg, M., and Raskob, G. (2001) PRO-TECT trial: oral SNAC-heparin vs enoxa-parin for preventing venous thromboem-bolism following total hip replacement. Blood100, 148–9a.

59. Kim, S.K., Lee, D.Y., Lee, E., Lee, Y.K.,Kim, C.Y., Moon, H.T., and Byun, Y. (2007)Absorption study of deoxycholic acid-heparinconjugate as a new form of oral anti-coagulant J Control Release 120, 4–10.

60. Mousa, S.A., Zhang, F., Aljada, A.,Chaturvedi, S., Takieddin, M., Zhang, H.,Chi, L., Castelli, M.C., Friedman, K., Gold-berg, M.M., and Linhardt, R.J. (2007) Phar-macokinetics and pharmacodynamics of oralheparin solid dosage form in healthy humansubjects J Clin Pharmacol 47, 1508–20.

61. Kelton, J.G. (2005) The pathophysiology ofheparin-induced thrombocytopenia: biologi-cal basis for treatment Chest 127, 9–20S.

62. Weitz, J.I., Hudoba, M., Massel, D.,Maraganore, J., and Hirsh, J. (1990) Clot-bound thrombin is protected from inhibi-tion by heparin-antithrombin III but is sus-ceptible to inactivation by antithrombin III-independent inhibitors J Clin Invest 86,385–91.

63. Mattsson, C., Menschik-Lundin, A., Nylan-der, S., Gyzander, E., and Deinum, J.(2001) Effect of different types of thrombininhibitors on thrombin/thrombomodulinmodulated activation of protein C in vitroThromb Res 104, 475–86.

64. Eriksson, B.I., Agnelli, G., Cohen, A.T.,Dahl, O.E., Lassen, M.R., Mouret, P.,Rosencher, N., Kalebo, P., Panfilov, S., Eskil-son, C., Andersson, M., and Freij, A. (2003)The direct thrombin inhibitor melagatranfollowed by oral ximelagatran compared withenoxaparin for the prevention of venousthromboembolism after total hip or kneereplacement: the EXPRESS study J ThrombHaemost 1, 2490–6.

65. Eriksson, B.I., Agnelli, G., Cohen, A.T.,Dahl, O.E., Mouret, P., Rosencher, N.,Eskilson, C., Nylander, I., Frison, L., andOgren, M. (2003) Direct thrombin inhibitormelagatran followed by oral ximelagatran incomparison with enoxaparin for preventionof venous thromboembolism after total hip


or knee replacement Thromb Haemost 89,288–96.

66. Fiessinger, J.N., Huisman, M.V., Davidson,B.L., Bounameaux, H., Francis, C.W., Eriks-son, H., Lundstrom, T., Berkowitz, S.D.,Nystrom, P., Thorsen, M., and Ginsberg,J.S. (2005) Ximelagatran vs low-molecular-weight heparin and warfarin for the treat-ment of deep vein thrombosis: a randomizedtrial JAMA 293, 681–9.

67. Francis, C.W., Berkowitz, S.D., Comp, P.C.,Lieberman, J.R., Ginsberg, J.S., Paiement,G., Peters, G.R., Roth, A.W., McElhattan,J., and Colwell, C.W., Jr. (2003) Compar-ison of ximelagatran with warfarin for theprevention of venous thromboembolism aftertotal knee replacement N Engl J Med 349,1703–12.

68. Francis, C.W., Davidson, B.L., Berkowitz,S.D., Lotke, P.A., Ginsberg, J.S., Lieber-man, J.R., Webster, A.K., Whipple, J.P.,Peters, G.R., and Colwell, C.W., Jr. (2002)Ximelagatran versus warfarin for the pre-vention of venous thromboembolism aftertotal knee arthroplasty. A randomized,double-blind trial Ann Intern Med 137,648–55.

69. Schulman, S., Wahlander, K., Lundstrom,T., Clason, S.B., and Eriksson, H. (2003)Secondary prevention of venous throm-boembolism with the oral direct thrombin

inhibitor ximelagatran N Engl J Med 349,1713–21.

70. Stangier, J., Rathgen, K., Gansser, D.,Kohlbrenner, V., and Stassen, J. (2001)Pharmacokinetics of BIBR 953 ZW, anovel low molecular weight direct throm-bin inhibitor in healthy volunteers. AbstractThromb Haemost 86, OC2347.

71. Eriksson, B.I., Dahl, O.E., Buller, H.R.,Hettiarachchi, R., Rosencher, N., Bravo,M.L., Ahnfelt, L., Piovella, F., Stangier, J.,Kalebo, P., and Reilly, P. (2005) A new oraldirect thrombin inhibitor, dabigatran etexi-late, compared with enoxaparin for preven-tion of thromboembolic events followingtotal hip or knee replacement: the BISTROII randomized trial J Thromb Haemost 3,103–11.

72. Eriksson, B.I., Dahl, O.E., Rosencher, N.,Kurth, A.A., van Dijk, C.N., Frostick, S.P.,Prins, M.H., Hettiarachchi, R., Hantel, S.,Schnee, J., and Buller, H.R. (2007) Dabi-gatran etexilate versus enoxaparin for pre-vention of venous thromboembolism aftertotal hip replacement: a randomised, double-blind, non-inferiority trial Lancet 370,949–56.

73. Coombe, S., Allen, G., and Kennedy, A.(2005) A phase I double-blind, ascendingdose study of an oral synthetic direct throm-bin inhibitor, TGN167 Blood 106, 530a.

Chapter 6

Oral Direct Factor Xa Inhibitors, with Special Emphasison Rivaroxaban

Shaker A. Mousa

Abstract

Rivaroxaban is a small-molecule, direct factor Xa inhibitor that is under investigation for the preventionand treatment of venous and arterial thrombosis. To date, oral anticoagulants have been limited largelyto vitamin K antagonists. Despite their remarkable benefits, vitamin K antagonists are limited by theirnarrow therapeutic window, the existence of multiple food and drug interactions, and the need for fre-quent monitoring and dose-adjustment. Rivaroxaban represents a potentially attractive alternative towarfarin, as it could enable simplified once-daily dosing, requires no therapeutic monitoring, and has alower potential for drug interactions. At present, the safety and efficacy of rivaroxaban for the prophylaxisand treatment of venous thromboembolism has been evaluated in phase-II and phase-III trials involvingover 24,000 patients. Rivaroxaban is also being evaluated for the treatment of pulmonary embolism, sec-ondary prevention after acute coronary syndromes, and the prevention of stroke and non-central nervoussystem embolism in patients with non-valvular atrial fibrillation. The need for new oral anticoagulants, thedevelopment and pharmacology of rivaroxaban, results of completed studies of rivaroxaban, and detailsof ongoing phase-II and phase-III trials with rivaroxaban are the subjects of this chapter.

Key words: Oral anticoagulant, rivaroxaban, BAY 59-7939, factor Xa inhibitor, venousthromboembolism.

1. Introduction

Arterial thrombosis, venous thrombosis, and subsequent throm-boembolism account for significant morbidity and mortalityworldwide. In the United States, more than 200,000 patientsdevelop venous thromboembolism (VTE) every year, and 30%of these patients die within 30 days (1). Despite administrationof current prophylaxis, 5–20% of all hip replacement surgeries


181

182 Mousa

are complicated by VTE (2). Furthermore, deep venous throm-bosis and pulmonary embolism are associated with a significanteconomic burden due to the costs of acute care as well as long-term costs associated with recurrent VTE and post-thromboticsyndrome (3). A problem growing at a faster rate is the increas-ing burden imposed by atrial fibrillation(AF)-associated throm-boembolism. AF leads to stasis of blood in the atria and forma-tion of thrombi that can leave the heart and embolize any vascu-lar bed, most seriously the cerebral circulation, leading to stroke.At present, 2.3 million Americans have AF, including 10% of allpatients 80 years and older. By 2050, an estimated 5.6 millionAmericans will have AF, increasing the risk of stroke fivefold, andultimately accounting for 15–20% of all strokes in the UnitedStates (4, 5). Furthermore, patients who have an AF-associatedstroke are twice as likely to remain bedridden as other stroke vic-tims (6). While there are several efficacious intravenous and sub-cutaneous alternatives for acute anticoagulation with a favorablebalance of effectiveness and safety in the setting of acute coronarysyndromes, deep venous thrombosis, or pulmonary embolism,there are few medications available for chronic, oral anticoagu-lation in AF. Oral anticoagulation thus far has been limited tovitamin K antagonists (VKAs), principally warfarin sodium.

1.1. Pharmacologyof Warfarin

Warfarin, first described by Karl Paul Link in 1940, is the currenttreatment of choice for the prevention of thromboembolism inpatients with AF (7, 8). By interfering with the cyclic intercon-version of vitamin K and vitamin K 2,3-epoxide, warfarin impairsthe γ-carboxylation of vitamin K-dependent proteins, includingimportant serine proteases in the coagulation cascade that requirevitamin K for their biologic activity (Factors II, VII, IX, and X).Warfarin is highly effective in preventing thromboembolic eventsin patients with AF. In 29 randomized trials involving more than28,000 patients, warfarin reduced the risk of stroke by 64% (8).Furthermore, warfarin was associated with a 26% (95% CI, 3–43%) reduction in all-cause mortality in randomized controlledtrials when compared to no anticoagulation therapy in patientswith AF (8). However, the benefits of warfarin in patients wellcontrolled with the agent might overestimate benefits comparedto the effects seen in warfarin-naive patients (9). Despite itseffectiveness in preventing stroke and non-central nervous sys-tem embolism, warfarin has significant limitations. It has a slowonset of action, narrow therapeutic window, and requires fre-quent monitoring due to marked inter-individual variation indrug metabolism, as well as multiple drug and food interactions.In fact, patients taking warfarin spend nearly a third of the timeoutside their target INR window (10). Unfortunately, patientswho spend more than 10% of their time outside of their targetINR window are more likely to suffer an ischemic stroke and

Oral Direct Factor Xa Inhibitors 183

mortality (10). Patients with the highest risk of stroke, i.e., elderlyindividuals with evidence of heart failure, hypertension, diabetes,or prior stroke, have the highest rates of major hemorrhage, witha discontinuation rate up to 26% in the first year of treatment(11). This is likely due to difficulty in maintaining patients withina therapeutic window.

Warfarin is a racemic compound, in which the S-warfarinenantiomer is a fivefold more potent vitamin K antagonist thanthe R-enantiomer (12). Each enantiomer exhibits unique interac-tions depending on an individual’s genetic background and con-comitant medications. Polymorphisms in cytochrome P450 C29(the enzyme responsible for metabolism of the S-isomer) havebeen associated with impaired hydroxylation of S-warfarin, lead-ing to low-warfarin dose requirements (13). S-warfarin clearanceis preferentially impaired by metronidazole and trimethoprim–sulfamethoxazole, while other medications such as amiodaroneinhibit both R- and S-warfarin clearance (14). As a result, pre-dicting the anticoagulation response with drug interactions isextremely complex.

Warfarin dosing is further complicated by dramatic inter-individual variability in warfarin metabolism. While the mean dailydose is 4.57 mg, more than 5% of patients require a daily dosegreater than 10 mg (15). Polymorphisms in the vitamin K recep-tor gene (VKORC1) have been associated with elevated warfarindose requirements (16). In addition to difficulties with dosageand frequent and expensive monitoring, warfarin is also associ-ated with significant adverse effects. Long-term warfarin use isassociated with a 25% increased risk of osteoporotic fractures (17).Patients on warfarin are at increased risk of life-threatening bleed-ing, including intracranial hemorrhage, which affected 1.8% ofpatients older than 75 in the Stroke Prevention in Atrial Fibril-lation II (SPAF II) study (18). It has also been observed thatthe risk of bleeding is highest in the first year of warfarin ther-apy (9). Finally, as with most therapeutics, there is the followingrisk/benefit paradox with respect to warfarin-associated bleed-ing: patients at highest risk of stroke, according to both age andcomorbidities, are also the patients with the highest risk of major,life-threatening hemorrhage (11, 19).

1.2. InvestigatingPotential Alternativesto Warfarin

Due to the limitations and risks associated with warfarin, less thanhalf of eligible patients are ultimately treated with warfarin forstroke prophylaxis (20, 21). Despite these limitations, warfarinhas remained the therapy of choice for prevention of thromboem-bolism in patients with AF since it became clinically available in1954. Recently, randomized trials have investigated alternatives towarfarin therapy. The Stroke Prevention using an Oral ThrombinInhibitor in Atrial Fibrillation (SPORTIF) trials compared the useof ximelagatran, a competitive inhibitor of human α-thrombin,

184 Mousa

with dose-adjusted warfarin (22, 23). At the time, ximelagatranrepresented a favorable alternative to warfarin as it had a pre-dictable anticoagulant profile with rapid oral absorption and arapid onset of action. Despite promising early results, a reviewof 6,948 patients treated with ximelagatran revealed transient ele-vations in alanine aminotransferase greater than 3 times the nor-mal upper limit (ALT > 3×ULN) in 7.9% of patients treated withximelagatran versus 1.2% in the comparator group (24). In addi-tion to three cases of fatal hepatotoxicity, the clinical trial resultsalso showed a troubling trend toward increased nonfatal myocar-dial infarction in patients treated with ximelagatran, especially inthose in whom it was recently discontinued. As a result, the USFood and Drug Administration did not approve ximelagatran forthe prevention of stroke and non-central nervous system (CNS)embolism in patients with AF (25). It was allowed on the marketin Europe, but was rapidly pulled by the manufacturer when afatal case of hepatic failure was attributed to the drug.

While warfarin reduces the risk of stroke by 45% (95% CI,29–57%) compared to antiplatelet monotherapy with aspirin, it isalso associated with a 70% increased risk of bleeding (26). TheACTIVE W study was a randomized trial designed to determinewhether dual antiplatelet therapy with aspirin and the thieno-phyridine clopidogrel was non-inferior to adjusted dose warfarintherapy (9). The trial was terminated early due to clear superiorityof oral anticoagulation with warfarin.

1.3. Safer and MoreEfficaciousAnticoagulation

The ideal oral anticoagulant would have a wide therapeuticwindow, rapid onset of action, minimal food and drug interac-tions, a short half-life allowing for quick termination in the eventof bleeding, a readily available antidote or reversal agent, andclear efficacy in large trials without adverse effects. When con-sidering candidates for potential new oral anticoagulants, atten-tion must be paid to the three temporal aspects of hemostasis,including (1) initiation, (2) amplification, and (3) termination(27). The prototypical anticoagulant would target the amplifica-tion phase without interfering with initiation or termination, inorder to allow some hemostasis in the event of tissue injury. Acti-vated factor Xa (FXa) is central to the coagulation cascade and isthe cornerstone of serine protease activity amplification. FX is avitamin K-dependent serine protease synthesized in the liver thatcan be activated by either the intrinsic or the extrinsic clottingcascade. Binding of FXa to activated FV in the presence of cal-cium on a phospholipid bilayer results in formation of the pro-thrombinase complex. FXa catalyzes the conversion of prothrom-bin (Factor II) to thrombin (Factor IIa) and is the rate-limitingstep in thrombin generation (Fig. 6.1) (28). One molecule ofFXa can catalyze the formation of over a thousand moleculesof thrombin (29). Thrombin potentiates clot formation by


Intrinsicpathway

Extrinsicpathway

FXa

Prothrombin Thrombin

FibrinogenFibrin clot

FXa

AT+

FX AT

Fig. 6.1. Rivaroxaban selectively inhibits FXa. Schematic representation of the mecha-nism of inhibition of FXa by rivaroxaban

up-regulating its own production through feedback activation offactors V, VII, VIII, and XI, and inducing platelet activation (30,31). Therefore, inhibition of FXa represents a potentially moreefficacious anticoagulation strategy than targeting all of the vita-min K-dependent clotting factors.

Inhibition of FXa can occur through direct binding to theFXa active site or indirectly through interaction with antithrom-bin. Direct FXa inhibitors have an advantage because they canbind both free FXa and FXa within the prothrombinase com-plex, and therefore penetrate the active thrombus to limit furtherthrombin generation.

2. ClinicalPharmacologyof Rivaroxaban

Rivaroxaban is an oxazolidinone derivative that binds to the activesite of FXa, leading to potent and selective inhibition of FXa(32). In animal models of both venous stasis and thrombosis, oralrivaroxaban inhibited FXa activity, leading to reduced thrombusformation and extension (33, 34). Rivaroxaban inhibits FXa activ-ity in a dose-dependent manner, accompanied by prolongation ofprothrombin time (PT) (Fig. 6.2). Phase-I data has shown that15 mg of rivaroxaban decreases FXa activity by 35% and increasesPT 1.4-fold over baseline values (35). The observed prolongationin PT correlated strongly with plasma rivaroxaban concentration(r = 0.935), with little inter-individual variability (36, 37). Thus,therapeutic monitoring might be possible through determinationof PT when necessary. In phase-I studies in healthy male vol-unteers, a single 30-mg dose of rivaroxaban inhibited thrombingeneration for greater than 24 h (38). Finally, rivaroxaban has nodirect effect on platelet aggregation.

186 Mousa

Plasma concentration of BAY 59-7939 (µg/l)

0 100 200 300 400 500 600

Pro

thro

mbi

n tim

e (s

)

0

10

20

30

40

PTModel

Fig. 6.2. Prothrombin time correlates strongly with plasma concentration of rivaroxaban (r = 0.958). This figure reprintedwith kind permission from Springer Science and Business Media (39).

2.1.Pharmacokinetics

The pharmacokinetic profile of rivaroxaban is consistent withrapid oral absorption and 80% bioavailability (36). The time topeak plasma concentration ranges from 2.5 to 4 h. After multi-ple doses, the drug half-life is 5–9 h in healthy volunteers, and9–13 h in the elderly (mean age 65) (39). There are no majoractive circulating metabolites of rivaroxaban. A portion of thedrug is excreted in the urine (1/3), and the remainder (2/3)is metabolized by the liver (36). Drug elimination demonstratesfirst-order kinetics and is impaired with advancing age, renal insuf-ficiency, and in the presence of strong cytochrome P450 3A4inhibitors (such as ketoconazole, macrolide antibiotics such asclarithromycin, and many protease inhibitors). In a phase-I studyof patients with renal insufficiency, subjects with severe renalimpairment (creatinine clearance <30 ml/min) experienced a 64%increase in serum drug concentration (p < 0.05 for AUC) and a144% prolongation of PT (p < 0.001) compared to control sub-jects (40). Accordingly, patients with a creatinine clearance lessthan 30 ml/min, significant liver disease, and those taking strongcytochrome P450 3A4 inducers or inhibitors are being excludedfrom phase-III trials of rivaroxaban.

While many anticoagulants are dose adjusted for extremes ofbody weight due to increased risks of bleeding, a phase-I random-ized placebo, parallel group study of rivaroxaban in patients withextremes of body weight (≤50 kg or >120 kg) demonstrated nochange in peak serum concentrations in those >120 kg, but didshow mild elevation (24%) in those ≤50 kg. This minor elevationwas associated with a small (15%) increase in PT, which was notconsidered clinically significant (41). Therefore, no dose adjust-ment is required for sex or body weight when dosing rivaroxaban.


While the available phase-I data have been largely limited tohealthy Caucasian males, studies examining rivaroxaban in race-and ethnicity-specific populations are ongoing.

2.2.PharmacodynamicInteractions

Rivaroxaban absorption is improved when taken with food;however, the drug can be administered on an empty stomach.The pharmacokinetics of the drug are unaffected by ranitidine oralteration of gastric pH with antacids in healthy male volunteers(42). Many patients who will require oral anticoagulation for AFor VTE prophylaxis often have risk factors for or documentedcoronary artery disease and require aspirin therapy. Rivaroxaban isbeing studied in patients with acute coronary syndromes, a patientpopulation treated with aspirin in addition to other antiplateletagents. In a randomized two-way crossover study (with healthymale subjects), antiplatelet therapy with aspirin did not alterthe pharmacokinetics or pharmacodynamics of rivaroxaban (asdetermined by bleeding and PTs). Furthermore, platelet aggre-gometry studies were unaffected by rivaroxaban (43). A similarphase-I, two-way crossover study demonstrated no clinically sig-nificant interaction between naproxen and rivaroxaban in healthysubjects (35).

Since rivaroxaban might have a role in stroke prophylaxis inthose with AF, 20 mg of rivaroxaban was co-administered in 20healthy male volunteers who also received 0.375 mg of digoxin.Drug exposure was not significantly different between patientsreceiving rivaroxaban alone and those who received rivaroxabanand digoxin. Based upon these results, there appears to be noapparent interaction between rivaroxaban and digoxin, suggest-ing that they can be prescribed together (44). Bridging withlow-molecular weight heparins (LMWHs) is common in patientsreceiving chronic oral anticoagulants. When given togetherwith rivaroxaban, enoxaparin resulted in additive inhibition ofFXa activity and prolongation of bleeding times; however, co-administration of LMWH and rivaroxaban has been demon-strated (45). Overall, compared to the VKAs and other cardio-vascular medications such as amiodarone, rivaroxaban has rela-tively low potential for substantial pharmacodynamic interactions,allowing for a wide range of concomitant pharmacotherapy.

2.3. Toxicityand Adverse Effects

Therapeutic anticoagulation always carries an attendant risk ofbleeding, either due to errors in dosing and administration,occult pathology such as gastric ulceration, unrecognized bleed-ing diatheses, or urgent and emergency medical procedures.Therefore, there is great interest in and need for neutralizingagents in the event of significant bleeding. To address this con-cern, investigators explored the use of recombinant activated fac-tor VII (rFVIIa) as a partial reversal agent for rivaroxaban. In a ratmodel of mesenteric hemorrhage, rFVIIa was administered after

188 Mousa

high-dose rivaroxaban (2 mg/kg). In the presence of high-dose,supratherapeutic rivaroxaban, 400 mcg/kg of rFVIIa reducedbleeding times by nearly 50% (46). rFVIIa also partially reversedthe prolongation of PT and partially restored total thrombinactivity, without affecting rivaroxaban-dependent FXa inhibition.Therefore, rFVIIa might be of use as an intravenous antidotefor major bleeding in patients taking rivaroxaban. While not yetinvestigated, factor VIII inhibitor bypassing activity (FEIBA R©),a vapor-heated lyophilized powder for reconstitution, also rep-resents a potential alternative for serious and/or life-threateningbleeding with rivaroxaban.

In 1,102 patients, rivaroxaban did not affect electrocardio-graphic parameters, including the QTc (47). This is important,as many patients with AF who might be candidates for rivaroxa-ban therapy will be taking antiarrhythmic drugs that are knownto prolong the QT interval, including dofetilide and sotalol.

Given the hepatotoxicity observed with ximelagatran (eventhough ximelagatran is a direct thrombin inhibitor, not a FXainhibitor), particular attention has been focused on liver func-tion surveillance in patients receiving rivaroxaban. In the recentOral Direct Factor Xa Inhibitor BAY 59-7939 in Patientswith Acute Symptomatic Deep-Vein Thrombosis (ODIXa-DVT)trial, there was no evidence of hepatotoxicity with long-term(3 months) administration of rivaroxaban for treatment of deepvenous thrombosis (DVT) (48). In the first 3 weeks of thistrial, patients randomized to rivaroxaban had a lower inci-dence of ALT >3×ULN (1.9–4.3% versus 21.6%) as comparedto those receiving enoxaparin. After 3 weeks however, the inci-dence of ALT >3×ULN was similar in both groups [1.9% (95%CI 0.8–3.6) versus 0.9% (95% CI 0.0–4.8)]. Rivaroxaban wasstopped early in three patients due to abnormal liver functiontests (two of these patients died, one from fulminant hepatitisB and one from carcinoma with hepatic metastasis) (48). In apooled analysis of 1,343 patients randomized in phase-II studiesof rivaroxaban for the prevention of post-operative VTE, therewas no difference in the incidence of ALT >3×ULN betweenrivaroxaban and enoxaparin (3.8–6.0% versus 7.7%) (47). Whilethere is no evidence of increased hepatotoxicity with rivaroxabanin multiple phase-II studies or in early reports of phase-III studiesof VTE prophylaxis, more long-term data are needed.

3. ClinicalIndications

As evidenced by the significant clinical and economic burdenimposed by VTE, including DVT and pulmonary embolism (PE),in both medical and surgical patients, the rising incidence of AF in


the rapidly expanding elderly population, and the global impactof ischemic heart disease, there are many potential patient pop-ulations and indications for novel, safe, and effective oral anti-coagulants such as rivaroxaban. Accordingly, a broad network ofclinical trials has been designed to evaluate the safety and efficacyof rivaroxaban in patients at risk for arterial and venous throm-bosis. In the sections to follow, we will review completed, ongo-ing, and planned clinical trials of rivaroxaban, according to clinicalindication.

4. Post-operativeThromboprophy-laxis

Proof of principle for the use of rivaroxaban in the preven-tion of VTE was first demonstrated in a phase-IIa study of 642patients undergoing total hip replacement (ODIXa-HIP). In thisopen-label, dose-escalation (12-fold dose range) study, patientswere randomized in a 3:1 ratio to rivaroxaban (2.5, 5, 10, 20,and 30 mg twice daily, or 30 mg once daily) or enoxaparin(40 mg daily) (49). Patients were treated until mandatory bilateralvenography was performed 5–9 days after surgery. The primaryefficacy endpoint (DVT, PE, or all-cause mortality) was not dif-ferent in those treated with rivaroxaban compared to enoxaparin(10.2–23.8% vs. 16.8%). There was a dose-dependent reduc-tion in major VTE (defined as proximal DVT, PE, VTE-relateddeath, Table 6.1) with rivaroxaban, which was accompanied bya dose-dependent (0–10.8%) increase in the incidence of majorpost-operative bleeding (P<0.001). While this study successfullydemonstrated proof of concept for the therapeutic efficacy ofrivaroxaban, it was limited by the open-label design and inade-quate power to compare efficacy between enoxaparin and indi-vidual rivaroxaban doses.

The ODIXa-KNEE trial was a randomized double-blind,double-dummy phase-IIb dose-ranging study in 621 patientsundergoing total knee replacement (50). There was no differ-ence between rivaroxaban (2.5, 5, 10, 20, 30 mg twice daily)and enoxaparin (30 mg twice daily) for the prevention of DVT,PE, and death. As previously observed, rivaroxaban was associatedwith a dose-dependent increase in major bleeding, but there wasno difference when compared to enoxaparin. In the companionODIXa-HIP2 trial, patients were randomized to the same esca-lating doses of rivaroxaban or enoxaparin 40 mg daily. ODIXa-HIP2, like ODIXa-KNEE, was a double-blind, double-dummytrial with treatment until mandatory venography 5–9 days aftersurgery. As observed in the ODIXa-KNEE trial, there was nodifference or dose effect on the composite endpoint (DVT, PE,

190 Mousa

Tabl

e6.

1Sa

fety

and

effic

acy

ofriv

arox

aban

:pha

se-I

Iast

udy

ofVT

Epr

ophy

laxi

sfo

llow

ing

tota

lhip

repl

acem

ent

Riva

roxa

ban

(dos

ean

dre

gim

en)

Enox

apar

inP-

valu

efo

rdo

se–r

espo

nse

Dos

e2.

5m

gbi

d5

mg

bid

10m

gbi

d20

mg

bid

30m

gbi

d30

mg

od40

mg

od

Prim

ary

endp

oint

(%D

VT

,PE

,or

all-

caus

em

orta

lity)

22.2

23.8

20.0

10.2

17.4

15.1

16.8

0.05

Maj

orV

TE

(%pr

oxim

alD

VT

,PE

,and

VT

E-r

elat

edde

ath)

11.1

7.9

3.6

04.

31.

44.

70.

01

Maj

orbl

eedi

ng(%

)0

2.5

2.9

6.5

10.8

4.5

0<0

.001

od=

once

ada

y,bi

d=

twic

ea

day


or death) between rivaroxaban and enoxaparin. Again, there wasa dose-dependent increase in major bleeding with rivaroxaban,but there was no difference compared to daily enoxaparin. Ina pooled analysis of both studies (ODIXa-KNEE and ODIXa-HIP2), which randomized a total of 1,343 patients (Table 6.2),there was a flat dose–response relationship for the preventionof total VTE with rivaroxaban (16.1–24.4% vs. 27.8% as com-pared to enoxaparin) (47). Not surprisingly, rivaroxaban wasassociated with a dose-dependent increase in incidence of majorbleeding (defined as bleeding leading to death, bleeding intoa critical organ such as the CNS, transfusion requirement >2units, >2 g/dl fall in hemoglobin, or re-operation). The inci-dence of major bleeding ranged from 0.9 to 7% with rivaroxa-ban as compared to 1.7% with enoxaparin. This dose-dependentrisk of bleeding remained significant, even after adjustment forage, gender, and study-specific bleeding rates. The majority ofmajor bleeding episodes, as expected, were confined to the oper-ative site. There was no significant difference in the risk of majorbleeding between enoxaparin and rivaroxaban (total daily doseof 5–20 mg). Accordingly, this dose range was suggested asthe optimal dosing for the prevention of VTE after orthopedicsurgery.

While ODIXa-KNEE and ODIXa-HIP2 compared twice-daily dosing of rivaroxaban, a randomized double-blind, double-dummy, active-comparator controlled trial of 873 patientsexamined once-daily rivaroxaban (over an eightfold dose rangecomprised of 5, 10, 20, 30, or 40 mg) compared to enoxaparin40 mg daily for the prevention of VTE after total hip replace-ment (ODIXa-OD-HIP) (51). As in prior trials, patients weretreated for 5–9 days after surgery and then underwent manda-tory bilateral venography. In this once-daily dosing trial, therewas a trend to significance in the dose-dependent reduction ofVTE in patients treated with rivaroxaban. Perhaps of greater rel-evance was the statistically significant dose-dependent reductionin major VTE (proximal DVT, PE, or death). There was a dose-dependent increase in the risk of major bleeding, however, thestrength of the relationship was less than previously observed inthe twice-daily dosing studies (2.3–5.1%, P = 0.039). Rivaroxa-ban was well tolerated in this trial, and no dose arm was stoppeddue to safety concerns. Based on the comparable efficacy betweendoses and the increase in major bleeding from 0.7 to 4.3% in the10 mg as compared to 20 mg groups, the authors recommended10 mg as the daily dose for future phase-III VTE preventiontrials.

All of these early trials led to the phase-III Regulation ofCoagulation in Major Orthopaedic Surgery Deducing the Risk of

192 Mousa

Tabl

e6.

2Ri

varo

xaba

ndo

sing

inth

rom

bopr

ophy

laxi

str

ials

Tria

lOD

IXa-

HIP2

and

KNEE

pool

ed(n

=1,3

43)

ODIX

a-OD

-HIP

(n=8

73)

RECO

RD3

(n=2

,531

)

Twic

e-da

ilydo

sing

(bid

)On

ce-d

aily

dosi

ngOn

ce-d

aily

dosi

ng

Dos

e(m

g)2.

55

1020

30P

-val

ue∗

40m

gen

ox.

510

2030

40P

-val

ue∗

40m

gen

ox.

1040

mg

enox

.P

-val

ue∗

Tot

alV

TE

(DV

T,

PE,a

ndal

l-ca

use

mor

talit

y)

21.6

22.9

16.1

24.4

19.3

0.39

27.8

14.9

10.6

8.5

13.5

6.4

0.08

525

.29.

618

.9<0

.001

Maj

orV

TE

(pro

xim

alD

VT

,PE

,or

VT

E-r

elat

edm

orta

lity)

32.

43.

13.

21.

10.

464.

58.

52.

70.

91.

91.

10.

007

2.8

12.

60.

01

Maj

orbl

eedi

ng0.

91.

32.

13.

97.

0<0

.001

1.7

2.3

0.7

4.3

4.9

5.1

0.03

91.

90.

60.

5N

S∗ P

-val

uefo

rdo

sing

tren

d;en

ox,e

noxp

arin

qd


DVT and PE (RECORD3) trial, which randomized over 2,500patients to treatment with either rivaroxaban 10 mg daily orenoxaparin 40 mg daily for the prevention of VTE after totalknee replacement. In this trial, enoxaparin was used in the man-ner recommended in the European package insert for the drug.Treatment with rivaroxaban led to a 49% relative risk reductionin the composite primary endpoint of DVT, PE, or death whencompared to enoxaparin (9.6 versus 18.9%, P<0.001) (52). Therewas also a 62% relative risk reduction in major VTE (the sec-ondary endpoint) with rivaroxaban (1 versus 2.6%, P = 0.01).This trial validated the selection of the 10 mg dose for prophy-laxis of VTE after surgery, as the incidence of major bleeding wascomparable to that with enoxaparin (0.6 versus 0.5%, P = NS). Insummary, RECORD3 demonstrated superior efficacy of rivarox-aban as compared to enoxaparin for the prevention of VTE,with similar bleeding rates among patients undergoing total kneereplacement.

5. Treatmentof VenousThrombosis

While the trials discussed above examined short-term anticoagu-lation with rivaroxaban, the ODIXa-DVT phase-II trial was thefirst to evaluate the use of rivaroxaban for treatment of knownthrombus and subsequent long-term anticoagulation. In ODIXa-DVT, patients were randomized in a parallel group design toanticoagulation with rivaroxaban (10, 20, 30 mg twice daily or40 mg once daily) or enoxaparin followed by oral VKA antago-nist therapy. The primary endpoint was improvement in throm-botic burden at day 21 (as determined by quantitative com-pression ultrasonography) without recurrent symptomatic VTEor VTE-related death. After 12 weeks of treatment, there wasno difference between rivaroxaban and enoxaparin for the pri-mary endpoint of thrombus reduction and recurrent VTE events(43.8–59.2% versus 45.9%), nor was there a dose–response rela-tionship with rivaroxaban (P = 0.67). In addition, there wasno dose–response between rivaroxaban and the primary safetyendpoint (major bleeding). This study demonstrated proof ofconcept that rivaroxaban could be used for the treatment ofexisting clots.

In the companion phase-II study, which evaluated once-daily dosing (EINSTEIN-DVT) with rivaroxaban, 543 patientswith symptomatic DVT without associated symptomatic PE wererandomized to rivaroxaban (20, 30, or 40 mg daily) or hep-arin/LMWH followed by a VKA for 12 weeks (53). The pri-mary outcome (deteriorating thrombus burden at 12 weeks

194 Mousa

as reflected by compression ultrasonography, perfusion lungscan, and recurrent symptomatic VTE) occurred in 5.4–6.6%of rivaroxaban patients as compared to 9.9% of heparin/VKApatients. There was no dose–response effect observed withrivaroxaban for the primary endpoint or for clinically relevantbleeding (primary safety endpoint) (2.9–7.5% versus 8.8% forheparin/VKA). The results of EINSTEIN-DVT, when viewed inthe context of the thromboprophylaxis trials and ODIXa-DVT,suggest that dose-dependent bleeding risk is most evident intwice-daily dose regimens as compared to once-daily dosing (seeTable 6.2).

The results of ODIXa-DVT and EINSTEIN-DVT, whichrandomized 1,156 patients, demonstrated that rivaroxaban caneffectively reduce thrombus burden in patients with DVT, withan acceptable safety profile. Nonetheless, phase-III studies usinghard clinical endpoints are required before long-term anticoagu-lation with rivaroxaban can be advocated in clinical practice. Atpresent, phase-III studies evaluating the use of rivaroxaban forthe treatment of PE (EINSTEIN-PE) and long-term anticoag-ulation in patients with DVT or PE who have already received6–12 months of oral rivaroxaban or VKA therapy (EINSTEIN-Extension) are currently recruiting patients.

6. OngoingPhase-II–IIIStudies inCardiovascularDisease

Cardiovascular disease remains the leading cause of death in theUnited States. Approximately one million Americans suffer amyocardial infarction each year. By the year 2050, it is estimatedthat over 5.6 million Americans will have AF, the most commonheart rhythm disorder encountered in clinical practice. There isgreat interest in the need for novel oral anticoagulants for theprevention of thromboembolic events in patients with AF as wellas the prevention of recurrent myocardial infarction and deathin patients with acute coronary syndromes (ACS). Rivaroxaban,with its associated safety and ease of dosing, is an attractive can-didate for anticoagulation for prevention of stroke in AF and sec-ondary prevention in patients with prior myocardial infarction.

6.1. Rivaroxabanin ACS

The Anti-Xa Therapy to Lower Cardiovascular Events in Additionto Aspirin with or without Thienopyridine Therapy in Subjectswith Acute Coronary Syndrome (ATLAS ACS TIMI 46) trialis a phase-II placebo-controlled randomized study designed toevaluate the safety of rivaroxaban in patients with recent ACS(www.clinicaltrials.gov; IdentifierNCT00402597). The trial willenroll patients between the ages of 18 and 75 who have symptoms


suggestive of ACS, a diagnosis of ST segment elevation or non-STsegment elevation myocardial infarction within the past 7 days,and at least one additional high-risk feature. Patients are beingrandomized to placebo, daily rivaroxaban, or twice-daily rivarox-aban. The trial will have two stages. The primary endpoint ofthe dose-escalation stage is significant TIMI bleeding. In the sub-sequent dose confirmation phase, the primary endpoint will be acomposite of major adverse cardiac events (including death, recur-rent myocardial infarction, stroke, or recurrent ischemia requiringrevascularization). Patient randomization will be stratified accord-ing to the presence or absence of thienopyridine treatment inorder to assess the risk benefit of anti-FXa activity with mono ordual antiplatelet therapy.

6.2. Rivaroxabanfor StrokeProphylaxisin Nonvalvular AF

The Rivaroxaban Once-Daily Oral Direct Factor Xa InhibitionCompared with Vitamin K Antagonism for Prevention of Strokeand Embolism Trial in Atrial Fibrillation (ROCKET-AF) trial isa prospective, randomized, double-blind, double-dummy, paral-lel group, multicenter, event-driven, non-inferiority study com-paring the efficacy and safety of once-daily oral rivaroxabanwith adjusted dose oral warfarin for the prevention of strokeand non-CNS embolism in patients with non-valvular AF. InROCKET-AF, patients are being randomized to rivaroxaban20 mg once-daily (15 mg if creatinine clearance is 30–49 ml/min)versus dose-adjusted warfarin. Blinded treatment will be main-tained through the use of a double-dummy system includingsham INRs in patients receiving rivaroxaban. The primary effi-cacy endpoint for the trial is a composite of stroke or non-CNS embolism. The objective of the primary efficacy analy-sis is to establish that rivaroxaban is not inferior to warfarin.The challenges of non-inferiority trials against warfarin are welldocumented (54, 55). Given the low frequency of stroke inpatients receiving dose-adjusted warfarin, a non-inferiority trialwill require a large sample size. ROCKET-AF will randomize atleast 14,000 patients in order to test the hypothesis.

The primary safety endpoint of ROCKET-AF is the compos-ite of major and non-major clinically relevant bleeding episodes.Secondary endpoints will include the individual components ofthe primary endpoints, in addition to myocardial infarction, dis-abling stroke, and all-cause mortality.

7. Expert Opinion

There is a clinical need for new oral anticoagulants. FXa representsan attractive pharmacologic target for new agents. While efficacyis paramount, so too is safety, given the morbidity and mortality

196 Mousa

associated with bleeding, especially in the predominantly elderlypopulation in whom oral anticoagulants are prescribed.

The drug development plan for rivaroxaban is aggressive,with simultaneous investigative programs spanning multiple indi-cations rather than a sequential approach. At present, over 24,000patients (12,000 are predominantly post-operative orthopedicpatients) have been evaluated in completed phase-II and III trialsof rivaroxaban for thromboprophylaxis and treatment of DVT. Bythe time all the currently enrolling trials have concluded, morethan 50,000 patients will have been evaluated in all randomizedcontrolled trials of rivaroxaban.

The advantages of rivaroxaban include the potential for once-daily dosing for all indications, no required dose-adjustment forbody weight, no known interactions with common cardiovascu-lar medications, a relatively safe pharmacodynamic profile withrespect to bleeding risk and hepatotoxicity, no clinically significantinteraction with aspirin, and the ability to bridge with LMWHwhen necessary. On the other hand, rivaroxaban is partially renallycleared and will require dose adjustment in those with grade-IIIchronic kidney disease, and is not being studied in patients witha creatinine clearance less than 30 ml/min. Since rivaroxabanmetabolism is affected by potent cytochrome P450 3A4 inhibitorssuch as ketoconazole, clarithromycin, and protease inhibitors, itsuse will be restricted in some special populations. Nonetheless,after extensive phase II, and now emerging phase III trial data,it appears that rivaroxaban is effective in preventing and treatingVTE with a bleeding risk comparable to other anticoagulants. Theresults of randomized trials evaluating rivaroxaban for the pre-vention of stroke and non-CNS embolism in AF and secondaryprevention of ACS are currently ongoing.

While there are several other oral anticoagulants in develop-ment, none have been evaluated as extensively and in as manypatients as rivaroxaban (Table 6.3). Rivaroxaban also has theadvantage that it can be dosed once a day, which has been shownto improve patient compliance and outcomes (56, 57). Despiteonce-daily dosing, rivaroxaban has a half-life that is considerablyshorter than other oral FXa inhibitors, which is an advantage inthe event of bleeding or an urgent need to discontinue anticoagu-lation. Unlike the direct thrombin inhibitor dabigatran, bioavail-ability of rivaroxaban is very good, and it has low potential fordrug–drug interactions, including medications which alter gas-tric pH that are taken chronically by 3% of the US population(58). Perhaps more importantly, rivaroxaban has been shown tohave no effect on platelet aggregation and its pharmacokineticprofile is unaffected by aspirin and other notable cardiovascularmedications, such as digoxin. Finally, after clinical investigation in


Table 6.3Some alternative oral anticoagulants in phase-III trials

Rivaroxaban(BAY 59-7939) Apixaban

Dabigatran(BIBR 1048)

Target Factor Xa Factor Xa Factor IIa (thrombin)

Dosing Once or twice daily Twice daily Once or twice dailyBioavailability (%) 80 24–88 4

Half-life (h) 5–9 and 11–13 in theelderly

9–14 8–17

Metabolism (%)RenalHepatic

33 (unchanged)67

2575

8020

Adjustments andinteractions

• Avoid with CrCl<30 ml/min

• Dose reduction forCrCl 30–49 ml/min

• Avoid strong CYP4503A4 inhibitors

• Avoid strongCYP450 3A4inhibitors

• Avoid protonpump inhibitors(↓ absorption by20–25%)

thousands of patients, rivaroxaban appears to have no significanthepatotoxicity and a bleeding risk comparable to other conven-tional anticoagulants.

8. Conclusion

Rivaroxaban is a novel, oral direct FXa inhibitor that does notrequire any cofactors. This therapeutic has potential use for severalclinical indications and disease states, including VTE prophylaxis,long-term treatment of DVT, PE, and the prevention of strokeand non-CNS embolism in patients with AF, and potentiallyACS. In order to understand the efficacy and safety of rivarox-aban across such a wide range of patient populations, severallarge simultaneous studies evaluating rivaroxaban in both venousand arterial thromboembolism are under way. Rivaroxaban mighthave its greatest impact in providing a much-needed and attractivealternative to warfarin. While formal cost-effective analyses arenot yet available, avoidance of the intensive, costly, and frequentmonitoring required with VKAs, as well as the potential to reduceadverse vascular events precipitated by the narrow therapeuticwindow of VKAs, should result in a significant improvement inquality of life and cost savings. While further data (particularly

198 Mousa

large phase-III trials) and caution are required, there is reason foroptimism. Rivaroxaban may very well be the long-awaited alter-native to warfarin.

References

1. Rosamond, W., Flegal, K., Friday, G.,Furie, K., Go, A., Greenlund, K., Haase, N.,Ho, M., Howard, V., Kissela, B., Kittner, S.,Lloyd-Jones, D., McDermott, M., Meigs, J.,Moy, C., Nichol, G., O’Donnell, C.J., Roger,V., Rumsfeld, J., Sorlie, P., Steinberger, J.,Thom, T., Wasserthiel-Smoller, S., Hong,Y., and for the American Heart AssociationStatistics Committee and Stroke Statistics, S.(2007) Heart disease and stroke statistics–2007 update: a report from the AmericanHeart Association Statistics Committee andStroke Statistics Subcommittee Circulation115, e69–171.

2. Caprini, J.A., Botteman, M.F., Stephens,J.M., Nadipelli, V., Ewing, M.M., Brandt, S.,Pashos, C.L., and Cohen, A.T. (2003) Eco-nomic burden of long-term complications ofdeep vein thrombosis after total hip replace-ment surgery in the United States ValueHealth 6, 59–74.

3. MacDougall, D.A., Feliu, A.L., Boccuzzi,S.J., and Lin, J. (2006) Economic burden ofdeep-vein thrombosis, pulmonary embolism,and post-thrombotic syndrome Am J HealthSyst Pharm 63, S5–15.

4. Go, A.S., Hylek, E.M., Phillips, K.A.,Chang, Y., Henault, L.E., Selby, J.V., andSinger, D.E. (2001) Prevalence of diagnosedatrial fibrillation in adults: national implica-tions for rhythm management and stroke pre-vention: the anticoagulation and risk factorsin atrial fibrillation (ATRIA) study JAMA285, 2370–5.

5. Wolf, P.A., Abbott, R.D., and Kannel, W.B.(1991) Atrial fibrillation as an independentrisk factor for stroke: the Framingham StudyStroke 22, 983–8.

6. Dulli, D.A., Stanko, H., and Levine, R.L.(2003) Atrial fibrillation is associated withsevere acute ischemic stroke Neuroepidemiol-ogy 22, 118–23.

7. Link, K. (1959) The discovery of dicumaroland its sequels Circulation 19, 97–107.

8. Hart, R.G., Pearce, L.A., Aguilar, M.I.(2007) Meta-analysis: antithrombotic ther-apy to prevent stroke in patients who havenonvalvular atrial fibrillation Ann Intern Med146, 857–67.

9. The ACTIVE Writing Group on behalf ofthe ACTIVE investigators (2006) Clopido-grel plus aspirin versus oral anticoagulation

for atrial fibrillation in the Atrial fibrillationClopidogrel Trial with Irbesartan for pre-vention of Vascular Events (ACTIVE W):a randomised controlled trial Lancet 367,1903–12.

10. Jones, M., McEwan, P., Morgan, C.L.,Peters, J.R., Goodfellow, J., and Currie,C.J. (2005) Evaluation of the pattern oftreatment, level of anticoagulation control,and outcome of treatment with warfarin inpatients with non-valvar atrial fibrillation: arecord linkage study in a large British popu-lation Heart 91, 472–7.

11. Hylek, E.M., Evans-Molina, C., Shea, C.,Henault, L.E., and Regan, S. (2007) Majorhemorrhage and tolerability of warfarin inthe first year of therapy among elderlypatients with atrial fibrillation Circulation115, 2689–96.

12. Breckenridge, A., Orme, M., Wesseling, H.,Lewis, R., and Gibbons, R. (1974) Phar-macokinetics and pharmacodynamics of theenantiomers of warfarin in man Clin Phar-macol Ther 15, 424–30.

13. Aithal, G.P., Day, C.P., Kesteven, P.J.L., andDaly, A.K. (1999) Association of polymor-phisms in the cytochrome P450 CYP2C9with warfarin dose requirement and risk ofbleeding complications Lancet 353, 717–9.

14. Hirsh, J., Dalen, J.E., and Guyatt, G.(2001) The sixth (2000) ACCP guidelinesfor antithrombotic therapy for preventionand treatment of thrombosis Chest 119,1S–2.

15. James, A.H., Britt, R.P., Raskino, C.L., andThompson, S.G. (1992) Factors affecting themaintenance dose of warfarin J Clin Pathol45, 704–6.

16. D’Andrea, G., D’Ambrosio, R.L.,Di Perna, P., Chetta, M., Santacroce, R.,Brancaccio, V., Grandone, E., andMargaglione, M. (2005) A polymorphismin the VKORC1 gene is associated withan interindividual variability in the dose-anticoagulant effect of warfarin Blood 105,645–9.

17. Gage, B.F., Birman-Deych, E., Radford,M.J., Nilasena, D.S., and Binder, E.F. (2006)Risk of osteoporotic fracture in elderlypatients taking warfarin: results from theNational Registry of Atrial Fibrillation 2 ArchIntern Med 166, 241–6.


18. Stroke Prevention in Atrial Fibrillation Inves-tigators (1994) Warfarin versus aspirin forprevention of thromboembolism in atrialfibrillation: stroke prevention in Atrial Fibril-lation II Study Lancet 343, 687–91.

19. Fang, M.C., Go, A.S., Hylek, E.M.,Chang, Y., Henault, L.E., Jensvold, N.G.,and Singer, D.E. (2006) Age and the risk ofwarfarin-associated hemorrhage: the antico-agulation and risk factors in atrial fibrillationstudy J Am Geriatrics Soc 54, 1231–6.

20. McCormick, D., Gurwitz, J.H., Goldberg,R.J., Becker, R., Tate, J.P., Elwell, A., andRadford, M.J. (2001) Prevalence and qualityof warfarin use for patients with atrial fibrilla-tion in the long-term care setting Arch InternMed 161, 2458–63.

21. Fang, M.C., Stafford, R.S., Ruskin, J.N.,and Singer, D.E. (2004) National trends inantiarrhythmic and antithrombotic medica-tion use in atrial fibrillation Arch Intern Med164, 55–60.

22. Olsson, S.B. (2003) Stroke prevention withthe oral direct thrombin inhibitor ximelaga-tran compared with warfarin in patients withnon-valvular atrial fibrillation (SPORTIFIII): randomised controlled trial Lancet 362,1691–8.

23. Albers, G.W., Diener, H.C., Frison, L.,Grind, M., Nevinson, M., Partridge, S.,Halperin, J.L., Horrow, J., Olsson, S.B.,Petersen, P., and Vahanian, A. (2005) Xime-lagatran vs warfarin for stroke prevention inpatients with nonvalvular atrial fibrillation: arandomized trial Jama 293, 690–8.

24. Lee, W.M., Larrey, D., Olsson, R., Lewis,J.H., Keisu, M., Auclert, L., and Sheth, S.(2005) Hepatic findings in long-term clin-ical trials of ximelagatran Drug Safety 28,351–70.

25. Boos, C.J., and Lip, G.Y.H. (2006) Ximela-gatran: an eulogy Thromb Res 118, 301–4.

26. van Walraven, C., Hart, R.G., Singer,D.E., Laupacis, A., Connolly, S., Petersen,P., Koudstaal, P.J., Chang, Y., andHellemons, B. (2002) Oral anticoagulantsvs aspirin in nonvalvular atrial fibrillation: anindividual patient meta-analysis JAMA 288,2441–8.

27. Rosenberg, R.D., and Aird, W.C. (1999)Vascular-bed – specific hemostasis and hyper-coagulable states N Engl J Med 340,1555–64.

28. Bauer, K.A. (2006) New anticoagulants: antiIIa vs anti Xa – is one better? J ThrombThrombolysis 21, 67–72.

29. Mann, K.G., Brummel, K., and Butenas, S.(2003) What is all that thrombin for?J Thromb Haemost 1, 1504–14.

30. Suzuki, K., Dahlback, B., and Stenflo, J.(1982) Thrombin-catalyzed activation ofhuman coagulation factor V J Biol Chem 257,6556–64.

31. Walsh, P.M.D. (2004) Platelet coagulation-protein interactions Sem Thromb Hemost461–71.

32. Roehrig, S., Straub, A., Pohlmann, J.,Lampe, T., Pernerstorfer, J., Schlemmer,K.H., Reinemer, P., and Perzborn, E. (2005)Discovery of the novel antithromboticagent 5-chloro-N-({(5S)-2-oxo-3- [4-(3-oxomorpholin-4-yl)phenyl]-1,3-oxazolidin-5-yl}methyl)thiophene- 2-carboxamide (BAY59-7939): an Oral, Direct Factor XaInhibitor J Med Chem 48, 5900–8.

33. Biemond, B.J., Perzborn, E., Friederich,P.W., Levi, M., Buetehorn, U., and Buller,H.R. (2007) Prevention and treatment ofexperimental thrombosis in rabbits withrivaroxaban (BAY 597939) – an Oral, DirectFactor Xa Inhibitor Thromb Haemost 97,471–7.

34. Perzborn, E., Strassburger, J., Wilmen, A.,Pohlmann, J., Roehrig, S., Schlemmer, K.H.,and Straub, A. (2005) In vitro and in vivostudies of the novel antithrombotic agentBAY 59-7939 – an Oral, Direct Factor XaInhibitor J Thromb Haemost 3, 514–21.

35. Kubitza, D., Becka, M., Mueck, W., andZuehlsdorf, M. (2007) Rivaroxaban (BAY59-7939) – an Oral, Direct Factor XaInhibitor – has no clinically relevant interac-tion with naproxen Br J Clin Pharmacol 63,469–76.

36. Kubitza, D., Becka, M., Voith, B.,Zuehlsdorf, M., and Wensing, G. (2005)Safety, pharmacodynamics, and pharmacoki-netics of single doses of BAY 59-7939, anOral, Direct Factor Xa Inhibitor Clin Phar-macol Ther 78, 412–21.

37. Mueck, W., Becka, M., Kubitza, D.,Voith, B., and Zuehlsdorf, M. (2007) Pop-ulation model of the pharmacokinetics andpharmacodynamics of rivaroxaban – an Oral,Direct Factor Xa Inhibitor – in healthy sub-jects Int J Clin Pharmacol Ther 45, 335–44.

38. Hader, S., Graff, J., Hentig, N.,Misselwitz, F., Kubitza, D., Zuelsdorf, M.,Wensing, G., Mueck, W., Becka, M., andBreddin, H.-K. (2003) Effects of BAY 59-7939, an Oral, Direct Factor Xa Inhibitor,on thrombin generation in healthy volun-teers Blood 102.

39. Kubitza, D., Becka, M., Wensing, G.,Voith, B., and Zuehlsdorf, M. (2005) Safety,pharmacodynamics, and pharmacokinetics ofBAY 59-7939—an Oral, Direct Factor XaInhibitor—after multiple dosing in healthy

200 Mousa

male subjects Eur J Clin Pharmacol 61,873–80.

40. Halabi, A., Maatouk, H., Klause, N.,Lufft, V., Kubitza, D., Zuehlsdorf, M.,Becka, M., Mueck, W., Schafers, R.,Wand, D., Philipp, T., and Bruck, H. (2006)Effects of renal impairment on the pharma-cology of rivaroxaban (BAY 59-7939) – Anoral, direct, factor Xa inhibitor ASH AnnuMeet Abs 108, Abs 913.

41. Kubitza, D., Becka, M., Zuehlsdorf, M., andMueck, W. (2007) Body weight has limitedinfluence on the safety, tolerability, pharma-cokinetics, or pharmacodynamics of rivaroxa-ban (BAY 59-7939) in healthy subjects J ClinPharmacol 47, 218–26.

42. Kubitza, D., Becka, M., Zuehlsdorf, M.,and Mueck, W. (2006) Effect of food, anantacid, and the H2 antagonist ranitidine onthe absorption of BAY 59-7939 (Rivaroxa-ban), an Oral, Direct Factor Xa Inhibitor,in healthy subjects J Clin Pharmacol 46,549–58.

43. Kubitza, D., Becka, M., Mueck, W., andZuehlsdorf, M. (2006) Safety, tolerability,pharmacodynamics, and pharmacokinetics ofrivaroxaban – an Oral, Direct Factor XaInhibitor – are not affected by aspirin J ClinPharmacol 46, 981–90.

44. Kubitza, D., Becka, M., Zuelsdorf, M., andMueck, W. (2006) No interaction betweenthe novel, Oral Direct Factor XA InhibitorBAY 59-7939 and digoxin. J Clin Pharmacol46, 11.

45. Kubitza, D., Becka, M., Voith, B., andZuehlsdorf, M. (2005) Effect of enoxaparinon the safety, tolerability, pharmacodynam-ics and pharmacokinetics of BAY 59-7939-an Oral, Direct Factor Xa Inhibitor J ThrombHaemost 3, 1704.

46. Tinel, H., Huetter, J., and Perzborn, E.(2006) Partial reversal of the anticoagulanteffect of high-dose rivaroxaban – an Oral,Direct Factor Xa Inhibitor – by recombinantFactor VIIa in rats Blood 108, 915.

47. Fisher, W.D., Eriksson, B.I., Bauer,K.A., Borris, L., Dahl, O.E., Gent,M., Haas, S., Homering, M., Huisman,M.V., Kakkar, A.K., Kalebo, P., Kwong,L.M., Misselwitz, F., and Turpie, A.G.(2007) Rivaroxaban for thromboprophy-laxis after orthopaedic surgery: pooled anal-ysis of two studies Thromb Haemost 97,931–7.

48. Agnelli, G., Gallus, A., Goldhaber,S.Z., Haas, S., Huisman, M.V., Hull,R.D., Kakkar, A.K., Misselwitz, F., andSchellong, S. (2007) Treatment of proximaldeep-vein thrombosis with the oral direct

factor Xa inhibitor rivaroxaban (BAY 59-7939): the ODIXa-DVT (Oral Direct FactorXa Inhibitor BAY 59-7939 in patients withacute smptomatic deep-vein thrombosis)study Circulation 116, 180–7.

49. Eriksson, B.I., Borris, L.C., Dahl, O.E.,Haas, S., Huisman, M.V., Kakkar, A.K.,Misselwitz, F., Muehlhofer, E., andKalebo, P. (2007) Dose-escalation studyof rivaroxaban (BAY 59-7939) – an Oral,Direct Factor Xa Inhibitor – for the pre-vention of venous thromboembolism inpatients undergoing total hip replacementThromb Res.

50. Turpie, A.G., Fisher, W.D., Bauer, K.A.,Kwong, L.M., Irwin, M.W., Kalebo, P.,Misselwitz, F., and Gent, M. (2005) BAY 59-7939: an Oral, Direct Factor Xa Inhibitor forthe prevention of venous thromboembolismin patients after total knee replacement.A phase II dose-ranging study J ThrombHaemost 3, 2479–86.

51. Eriksson, B.I., Borris, L.C., Dahl, O.E.,Haas, S., Huisman, M.V., Kakkar, A.K.,Muehlhofer, E., Dierig, C., Misselwitz,F., and Kalebo, P. (2006) A once-daily,Oral, Direct Factor Xa Inhibitor, rivaroxa-ban (BAY 59-7939), for thromboprophylaxisafter total hip replacement Circulation 114,2374–81.

52. Lassen, M., Turpie, A.G., Rosencher, N.,Borris, L., Ageno, W., Lieberman, J.,Bandel, T., and Misselwitz, F. Rivaroxaban:an Oral, Direct Factor Xa Inhibitor for theprevention of venous thromboembolism intotal knee replacement surgery – results ofthe RECORD3 study. In: XXIst Congress ofthe International Society on Thrombosis andHaemostasis; 2007; Geneva, Switzerland;2007.

53. Büller, H., and Agnelli, G. (2006) Once- ortwice-daily rivaroxaban for the treatment ofproximal deep vein thrombosis: similar effi-cacy and safety to standard therapy in dose-ranging studies Blood 108, 572.

54. Connolly, S.J., Eikelboom, J., O’Donnell, M.,Pogue, J., and Yusuf, S. (2007) Challengesof establishing new antithrombotic thera-pies in atrial fibrillation Circulation 116,449–55.

55. Kaul, S., Diamond, G.A., and Weintraub,W.S. (2005) Trials and tribulations of non-inferiority: the ximelagatran experience J AmColl Cardiol 46, 1986–95.

56. Claxton, A.J., Cramer, J., and Pierce, C.(2001) A systematic review of the asso-ciations between dose regimens andmedication compliance Clin Ther 23,1296–310.


57. Richter, A., Anton, S.E., Koch, P.,and Dennett, S.L. (2003) The impactof reducing dose frequency on healthoutcomes Clin Ther 25, 2307–35;discussion 6.

58. Jacobson, B.C., Ferris, T.G., Shea, T.L.,Mahlis, E.M., Lee, T.H., and Wang, T.C.(2003) Who is using chronic acid suppres-sion therapy and why? Am J Gastroenterol 98,51–8.

Chapter 7

Antiplatelet Therapies: Drug Interactionsin the Management of Vascular Disorders

Shaker A. Mousa

Abstract

Antiplatelet drugs represent a key class of drugs that are of proven value in arterial thromboembolicdisorders. There is a need for effective, safe antiplatelet agents or their combinations to provide pre-dictable therapeutic benefit, dosage flexibility, and unique pharmacologic profiles, such as rapid onset inacute thrombotic states, as well as sustained antiplatelet effects in chronic platelet-activating states (e.g.,post-stent placement). Aspirin, clopidogrel, or their combination have shown improved clinical outcomesin certain unique settings, and the search for additional antiplatelet agents is ongoing. Current studiessuggest that combination antiplatelet therapy with existing agents is best considered a use-adapted strat-egy, with the greatest clinical benefit of combination therapy realized in acute, platelet-activating, andprothrombotic states.

Key words: Acute coronary syndrome, clopidogrel, aspirin, myocardial infarction, antiplatelets,combination therapy, ADP receptor antagonist, percutaneous coronary intervention.

1. Introduction

Platelets play a key role in arterial thrombosis, where plateletactivation and aggregation are the proximate events associatedwith acute coronary syndrome (ACS), stroke, and peripheralartery disease (PAD) (1–6). Platelet adhesion to injured vas-cular endothelium leads to platelet activation, which is furtheramplified by various platelet agonists, including arachidonic acid,ADP, thrombin, serotonin, and collagen (5–7). Aspirin was thefirst antiplatelet drug to provide insight into the role of plateletsin health and disease (6, 8). Since its development in 1899and the subsequent elucidation of its mechanism of action,


203

204 Mousa

thromboxane A2 blockade has become one of the most well-known antiplatelet paradigms (8). Aspirin inhibits the forma-tion of thromboxane by irreversibly inhibiting COX-1 via acety-lation of the serine-529 peptide, the requisite site for the for-mation of thromboxane from arachidonic acid. It was not untilthe late 1970s that aspirin was widely recognized for its thera-peutic benefit in cardiovascular disease (2, 9). Aspirin is knownto reduce the risk of myocardial infarction (MI), lower the riskof developing ischemic stroke, and decrease mortality in patientswith vascular disease (10). Aspirin is presently recommended forthe management of acute MI and unstable angina and is aneffective therapeutic agent for secondary prevention of ischemicevents (11).

The development of the first ADP receptor antagonist, ticlo-pidine, followed the recognized clinical benefit of aspirin (12).Ticlopidine also demonstrated antithrombotic benefits but wasassociated with a high incidence of neutropenia, aplastic anemia,and thrombotic thrombocytopenic purpura (13). The discoveryof a second thienopyridine ADP receptor antagonist, clopido-grel, confirmed the promise of antiplatelet therapy via alterationof the platelet functional pathway. Clopidogrel was comparableto ticlopidine in terms of efficacy, but had a better safety pro-file (14, 15) and was associated with a lower incidence of neu-tropenia and thrombotic thrombocytopenic purpura (16). Withthe development of various antiplatelet therapies, considerationwas given to the therapeutic strategy of combining differentantiplatelet agents (17, 18). One of the earliest examples of theclinical success of such dual antiplatelet therapy was Aggrenox R©,a combination of a low-dose aspirin with dipyridamole for strokeprevention (19).

While aspirin continues to be used routinely for the manage-ment of acute MI, unstable angina, and secondary preventionof ischemic events, dual antiplatelet therapy has been tested inseveral different clinical populations to achieve optimal efficacywith minimal adverse affects. For example, aspirin has been com-bined with clopidogrel in different doses and at different dura-tions to assess its effectiveness as compared to aspirin alone (20)(Table 7.1). Improved efficacy has not always been the case, how-ever, and in some cases, dual antiplatelet therapy carries a signif-icant hemorrhagic risk (21). When contemplating a combinationtherapy, there are several factors that must be carefully weighedagainst the benefits. Each agent should individually demonstratesignificant clinical benefits, and the combination of agents shouldhave a greater benefit/risk ratio than either one alone. Ideally,each agent should act at a different mechanistic level and with dif-ferent capacities. When combining the agents at different doses,one should be able to demonstrate improved efficacy without acompromise in safety.

Antiplatelet Therapies: Drug Interactions in the Management of Vascular Disorders 205

Tabl

e7.

1Su

mm

ary

ofco

mbi

natio

nth

erap

ies

with

clop

idog

relp

lus

aspi

rinve

rsus

aspi

rinal

one

indi

ffere

ntpa

tient

popu

latio

ns

Tria

lSt

udy

popu

latio

nDe

sign

Outc

ome

CL

AR

ITY-

TIM

I3,

491

patie

nts,

acut

ese

ttin

g:pa

tient

sha

dex

peri

ence

dis

chem

iaw

ithin

12h

prio

rto

trea

tmen

t

Mul

ti-ce

nter

,ra

ndom

ized

,do

uble

-blin

d,pl

aceb

o-co

ntro

lled.

All

patie

nts

rece

ived

aspi

rin

and

wer

eth

enra

ndom

ized

tore

ceiv

eei

ther

clop

idog

relo

rpl

aceb

o.D

urat

ion:

30da

ys

Sign

ifica

ntre

duct

ion

(36%

)of

com

pos-

itede

ath,

recu

rren

tM

I,an

dne

edfo

rre

vasc

ular

izat

ion

was

seen

with

com

bina

tion

ther

apy

asco

mpa

red

toas

piri

nal

one.

No

sign

ifica

ntin

crea

sein

maj

orbl

eedi

ngw

asob

serv

edin

the

com

bina

tion

grou

p

CO

MM

IT45

,852

patie

nts,

acut

ese

ttin

g:pa

tient

ssu

spec

ted

ofac

ute

MI

Mul

ti-ce

nter

,ra

ndom

ized

,do

uble

-blin

d,pl

aceb

o-co

ntro

lled.

All

patie

nts

rece

ived

aspi

rin

and

wer

eth

enra

ndom

ized

toei

ther

clop

idog

relo

rpl

aceb

o.D

urat

ion:

disc

harg

eor

4w

eeks

Sign

ifica

ntre

duct

ion

(9%

)of

deat

han

dre

infa

rctio

nw

ere

seen

with

com

bina

-tio

nth

erap

yas

com

pare

dto

aspi

rin

alon

e.C

ombi

natio

nth

erap

yw

asno

tbe

nefic

iali

nre

duci

ngst

roke

CH

AR

ISM

A15

,603

patie

nts,

non-

acut

ese

ttin

g:al

lpa

tient

sw

ere

athi

ghri

skfo

rat

hero

thro

mbi

cev

ents

.Pa

tient

sw

ere

furt

her

defin

edas

eith

ersy

mpt

omat

ic(t

hose

with

docu

-m

ente

dca

rdio

vasc

ular

dise

ase)

oras

ympt

omat

ic(t

hose

who

only

disp

laye

dri

skfa

ctor

s)

Mul

ti-ce

nter

,ra

ndom

ized

,do

uble

-blin

d,pl

aceb

o-co

ntro

lled.

All

patie

nts

rece

ived

aspi

rin,

and

wer

eth

enra

ndom

ized

tore

ceiv

eei

ther

clop

idog

relo

rpl

aceb

o.D

urat

ion:

28m

onth

s

Patie

nts

inth

esy

mpt

omat

icsu

bgro

up:

risk

rate

redu

ctio

nof

1%in

prim

ary

end

poin

tsw

ithco

mbi

natio

nth

er-

apy

asco

mpa

red

toas

piri

nal

one.

Patie

nts

inas

ympt

omat

icsu

bgro

up:

1%in

crea

sein

the

rate

ofpr

imar

yen

dpo

ints

whe

ngi

ven

com

bina

tion

ther

-ap

yas

com

pare

dto

aspi

rin

alon

e.N

oin

crea

sein

maj

orbl

eedi

ngw

asse

enw

ithth

ead

ditio

nof

clop

idog

rel

206 Mousa

Tabl

e7.

1Co

ntin

ued

Tria

lSt

udy

popu

latio

nDe

sign

Outc

ome

CU

RE

and

PCI-

CU

RE

12,5

62pa

tient

s,ac

ute

sett

ing:

patie

nts

had

rece

ived

CA

BG

and

PCI

inte

rven

tions

and

had

sym

ptom

sin

dica

tive

ofA

CS

with

evid

ence

ofis

chem

ia

Mul

ti-ce

nter

,ra

ndom

ized

,do

uble

-blin

d,pl

aceb

o-co

ntro

lled.

All

patie

nts

rece

ived

aspi

rin,

and

wer

eth

enra

ndom

ized

tore

ceiv

eei

ther

clop

idog

relo

rpl

aceb

oD

urat

ion:

1ye

ar

CU

RE

:si

gnifi

cant

redu

ctio

n(2

%)

inth

epr

imar

yen

dpo

int

com

posi

teof

card

iova

scul

arde

ath,

MI,

orst

roke

was

seen

with

com

bina

tion

ther

apy

asco

mpa

red

toas

piri

nal

one

PCI-

CU

RE

sub-

anal

ysis

:pa

tient

sw

hore

ceiv

edPC

Iin

terv

entio

nde

mon

-st

rate

da

larg

erde

gree

ofre

lativ

eri

skre

duct

ion

(4%

)w

ithno

sign

ifica

ntex

cess

inm

ajor

blee

ding

MA

TC

H7,

599

patie

nts,

non-

acut

ese

ttin

g:pa

tient

sha

dex

peri

ence

dis

chem

icst

roke

orT

IAw

ithin

the

past

3m

onth

san

ddi

spla

yed

addi

tiona

lri

skfa

ctor

s

Mul

ti-ce

nter

,ra

ndom

ized

,do

uble

-blin

d,pl

aceb

o-co

ntro

lled.

Patie

nts

rece

ivin

gcl

opid

ogre

lw

ere

rand

omiz

edto

rece

ive

eith

eras

piri

nor

plac

ebo

Dur

atio

n:18

mon

ths

No

sign

ifica

ntdi

ffer

ence

was

obse

rved

inth

ere

duct

ion

ofM

I,va

scul

arde

ath,

orac

ute

isch

emic

even

tsw

ithco

mbi

natio

nth

erap

yas

com

pare

dto

aspi

rin

alon

e.C

ombi

natio

nth

er-

apy

grou

pex

peri

ence

da

sign

ifica

ntly

high

erra

teof

maj

orbl

eedi

ngC

AR

ESS

107

patie

nts,

acut

ese

ttin

g:pa

tient

sw

ere

rece

ntly

sym

ptom

atic

with

>50%

caro

tidst

enos

is,

and

ipsi

-la

tera

lca

rotid

terr

itory

TIA

orst

roke

with

in3

mon

ths

prio

rto

trea

tmen

t

Mul

ti-ce

nter

,ra

ndom

ized

,do

uble

-blin

d,pl

aceb

o-co

ntro

lled.

All

patie

nts

rece

ived

aspi

rin

and

wer

eth

enra

ndom

ized

tore

ceiv

eei

ther

clop

idog

rel

orpl

aceb

o.M

icro

embo

liw

ere

scre

ened

at7

days

with

tran

scra

nial

Dop

pler

.Dur

atio

n:7

days

Sign

ifica

ntre

duct

ion

(39%

)in

the

pri-

mar

yen

dpo

intw

asse

enw

ithco

mbi

-na

tion

ther

apy

asco

mpa

red

toas

piri

nal

one


2. CombinationAntiplatelets:Issues andPerspective

2.1. Heparin andPlatelet GPIIa/IIIaAntagonists

The most recent forms of antiplatelet therapy focus on fibrinogendisplacement in existing thrombi to prevent further plateletcrosslinking and thrombosis via the platelet GPIIb/IIIa complex(22–25). Abciximab, tirofiban, and eptifibatide are examples ofGPIIb/IIIa receptor inhibitors that have shown clinical benefitsin various trials (26–31) and have been incorporated into cur-rent American College of Cardiology/American Heart Associa-tion (ACC/AHA) treatment guidelines for unstable angina/non-ST elevation MI (32).

There is a wealth of clinical experience with the use of hep-arin and intravenous platelet GPIIb/IIIa antagonists in ACS andcombinations of intravenous GPIIb/IIIa antagonists with throm-bolytics. The potential clinical benefit of the platelet GPIIb/IIIaantagonist abciximab in ACS was demonstrated in the pivotalEPIC and EPILOG trials. In the EPIC trial, there was signif-icant excess bleeding that occurred when unfractionated hep-arin (UFH) was used in its full dose with abciximab, leading tothe EPILOG trial, where a reduced dose of heparin was used.The lower heparin dose led to an improved safety profile with-out compromising the efficacy observed in the EPIC trial. Simi-larly, the PRISM and the PRISM-PLUS trials evaluated whetheradministration of aspirin plus tirofiban (PRISM) or aspirin,heparin, and tirofiban (PRISM-PLUS) would improve clinicaloutcomes in the management of unstable angina. Tirofiban plusheparin was significantly more effective than tirofiban withoutheparin in reducing the incidence of death, MI, or refractoryischemia. Surprisingly, there was increased mortality with intra-venous tirofiban without heparin, similar to that observed in thevarious oral GPIIb/IIIa antagonist trials, suggesting the crit-ical need for heparin or anticoagulants with intravenous andperhaps oral GPIIb/IIIa antagonists. The interactions betweenplatelet GPIIb/IIIa receptor antagonists and heparin or low-molecular weight heparin (LMWH) could have tremendous clin-ical implications. LMWH may enhance the antiplatelet activ-ity of GPIIb/IIIa antagonists by inhibiting thrombin, FXa, andother coagulation factors, along with inducing the vascular releaseof tissue factor pathway inhibitor (TFPI). At the same time,GPIIb/IIIa antagonists may potentiate the anticoagulant actionof LMWH by blocking fibrinogen binding and aggregation,regardless of the activating stimulus and the down-regulationof pro-coagulant activity on the platelet surface. The combina-tion of reduced dose LMWH or anticoagulant with a reduced

208 Mousa

dose GPIIb/IIIa antagonist and the combination of GPIIb/IIIaantagonist with thrombolytic may result in a higher therapeuticindex in various thrombotic disorders and beyond.

2.2. Aspirinand Clopidogrel

This segment will focus on the rationale and current clinical statusof dual antiplatelet therapy consisting of aspirin and clopidogrel(versus monotherapy with aspirin alone) in arterial vascular dis-ease. The CLARITY-TIMI and COMMIT trials provided thebasis for the use of aspirin plus clopidogrel for patients suffer-ing from acute MI. In contrast, however, the CHARISMA trialdemonstrated that the combination was not justified in stablehigh-risk patients with documented coronary disease, cerebrovas-cular disease, or symptomatic PAD. In two additional cardiovas-cular studies, the CURE and the PCI-CURE trials, the bene-fit of the drug combination was demonstrated in patients whounderwent percutaneous coronary intervention (PCI) and coro-nary artery bypass graft (CABG). Finally, we will discuss twofocused cerebrovascular studies, the CARESS and MATCH tri-als. The CARESS trial demonstrated the clinical benefit of thecombination in an acute clinical setting. However, the combina-tion proved to be ineffective in the longer term MATCH trial.It appears from these large clinical trials that the risk benefit ofdual antiplatelet therapy with aspirin and clopidogrel is justified inhigh-risk symptomatic patients but not in asymptomatic patients.The exact dose regimen and duration of combination therapyawait definition and require careful assessment to optimize theanticoagulant benefit while minimizing the hemorrhagic risk.

2.2.1. Limitationsof Combined Therapy

Aspirin established the clinical benefit of antiplatelet agents, andclopidogrel exhibited augmented efficacy when combined withaspirin in certain, but not all clinical settings. However, theseagents are not without shortcomings, and their limitations area necessary backdrop to the studies discussed. Aspirin resis-tance, a phenomenon described clinically as a lack of desiredresponse while the patient is receiving aspirin, is estimated toaffect ≥20% of the population. Either on the basis of geneticpolymorphism, noncompliance, or concomitant administrationof a competing medication (e.g., an NSAID), aspirin resistanceremains an uncontrolled variable without a laboratory definition(33, 34).

Clopidogrel, a prodrug, suffers from delayed onset, a limi-tation that can be partially overcome by dose loading. The useof this strategy in some of the studies discussed below is noted.Other thienopyridines, such as prasugrel, which do not have thislimitation, are currently in phase-III clinical trials (35). There isalso evidence that the dose benefit of clopidogrel co-therapy maybe disease related, as in the case of type II diabetes, in which60% of the patients demonstrated platelet reactivity after receiving


twice the standard maintenance dose of typical clopidogreltherapy (36). As with aspirin, clopidogrel noncompliance is also aserious clinical confounding element, and in the presence of drug-eluting stents, is strongly associated with subsequent mortality.Regardless of the status of development of a given antiplateletagent or combination therapy, full clinical benefit can only be real-ized by parallel strategies in patient education to ensure medica-tion compliance (37). In the case of unstable angina and non-ST-segment elevation MI, clinical guides detailing the use of aspirinand clopidogrel have been established.

As mentioned earlier, a great concern of combination therapyis the increased risk for bleeding. In the MATCH (Managementof Atherothrombosis with Clopidogrel in High-Risk Patients withRecent Transient Ischemic Attack or Stroke) trial (see Section2.2.6), increased risk of bleeding, possibly associated with theduration of therapy, proved to be a significant issue.

Finally, discussion of current dual antiplatelet therapy wouldbe incomplete without the mention of combined therapy inthe prevention of stent thrombosis, in which dual therapy isconsidered the standard of care. Combined antiplatelet therapyreduces stent thrombosis, and current ACC/AHA guidelines onPCI recommend clopidogrel plus aspirin (38). This is an areaof great controversy regarding dose and duration of dual ther-apy, as guidelines were based on studies done with bare metalstents. In the current therapeutic era, most stents used are of thedrug-eluting type in which delayed neointimal revascularizationpresents a risk for late (> 6 months) stent thrombosis.

2.2.2. The CLARITY-TIMIStudy

The CLARITY-TIMI (Clopidogrel as Adjunctive ReperfusionTherapy-Thrombolysis in Myocardial Infarction 28) trial, a largetrial that involved 28 centers, compared the use of clopidogrelwith aspirin to aspirin alone (39). In this study, 3,491 patientswho experienced ischemia lasting >20 min within the prior 12 hand associated ST-segment elevation or new-onset left-bundlebranch block were scheduled to receive a fibrinolytic agent,aspirin, and heparin. The subjects were randomized to receiveclopidogrel or placebo and were followed for 30 days. The com-posite end point was occlusion of the infarct-related artery ordeath from any cause prior to angiography. The primary safetyend point was major hemorrhage defined by TIMI criteria afterangiography.

Patients in the clopidogrel-treatment group had a compos-ite reduction in recurrence of thrombosis in the infarct-relatedartery, MI, or death with a risk reduction of 36% (P < 0.001)at 2–8 days and a 20% reduction at 30 days. There was no sig-nificant increase in major bleeding between the two groups orany of the subgroups. The CLARITY-TIMI study showed thecombination of clopidogrel and aspirin to be superior to aspirin

210 Mousa

alone in decreasing the rate of death and recurrent thrombosis ininfarct-related arteries.

2.2.3. The COMMITStudy

In the COMMIT (Clopidogrel and Metoprolol in MyocardialInfarction Trial) trial (40), 45,852 patients suspected of acuteMI received aspirin (162 mg) and were randomized to aspirinplus placebo or aspirin plus clopidogrel. Patients also receivedmetoprolol or placebo. Patients did not receive a loading doseof clopidogrel and were monitored for 28 days or until discharge,whichever came first. Primary end points included a composite ofdeath, reinfarction, stroke, or death from any cause.

Primary end points were reached in 9.2% of the clopidogrelgroup versus 10.1% in the placebo group, and there was a 0.9%risk reduction of death, reinfarction, and stroke (P < 0.002) infavor of the clopidogrel group. There was also a significant (7%)reduction in death of any cause. This was accomplished with nosignificant excess bleeding in the clopidogrel-treated patients or inpatients aged >70 years, or those given fibrinolytic therapy. Basedon this study, it was concluded that routine use of clopidogrelplus aspirin in patients with acute MI safely reduces mortality andmajor vascular events in a large range of patients, including thoseaged >70 years.

2.2.4. The CHARISMAStudy

The CHARISMA (Clopidogrel for High Artherothrombotic Riskand Ischemic Stabilization, Management, and Avoidance) trialwas a 28-month trial evaluating antiplatelet combination ther-apy versus aspirin monotherapy for both primary and secondaryprevention of atherothrombotic events in 15,603 stable patients(41). All patients were categorized as high risk for atherothrom-botic events and were randomized to clopidogrel 75 mg/dayplus low-dose aspirin (75–162 mg/day) or low-dose aspirinplus placebo. The efficacy end point was a composite of MI,stroke, or death from cardiovascular disease. Patients were cat-egorized into two subgroups: a symptomatic subgroup (12,153patients), composed of those with documented cardiovascular dis-ease (remote MI, stroke or symptomatic PAD) and an asymp-tomatic group (3,284 patients), who were enrolled with multipleatherothrombotic risk factors but without established atheroscle-rosis. The primary safety end point was any event of severe bleed-ing based on GUSTO (Global Utilization of Streptokinase andTissue Plasminogen Activator for Occluded Coronary Arteries)criteria.

Overall, treatment was discontinued in 20.4% of patientsin the clopidogrel group and in 18.2% of the placebo group(P< 0.001) due to adverse events. The primary end pointoccurred in 6.8% of the clopidogrel group and in 7.3% of theplacebo group, which was not significant (P< 0.22). Severe bleed-ing occurred in 1.7% of the clopidogrel group and in 1.3% of the


placebo group (P< 0.09). The relative risk reduction in the symp-tomatic group received clopidogrel (6.9%) versus that placebo(7.9%) was 0.88 (P = 0.046). The results of the CHARISMAstudy indicated that only in the selected symptomatic group wasthere a suggestion of risk reduction that outweighed the risk ofbleeding, and in the asymptomatic group with multiple risk fac-tors, there was a suggestion of harm. Clopidogrel plus aspirin wasnot more effective than aspirin alone in reducing the rate of deathfrom MI, stroke, or death from cardiovascular disease in stablepatients in this long-term study.

2.2.5. The CURE andPCI-CURE Studies

The CURE (Clopidogrel in Unstable Angina to Prevent Recur-rent Events) trial evaluated the benefits and risks of clopidogrelplus aspirin versus aspirin alone in patients with non-ST elevationACS (42). A total of 12,562 patients were included in the trial;2,072 underwent CABG intervention, 2,658 underwent a PCI,and 2,658 were managed medically. Patients were included in thestudy if they had symptoms indicative of ACS within the preced-ing 24 h without ST-segment elevations, supporting evidence ofischemia from their most recent electrocardiogram, and elevatedconcentrations of cardiac enzymes (including troponin) that wereat least twice the reference range. All patients received aspirin75–325 mg/day and were then randomized to receive clopido-grel 75 mg/day with a loading of 300 mg or placebo for 3–12months. The primary outcome was a composite of cardiovascu-lar death, MI, or stroke among patients who underwent CABG,PCI, or medical therapy.

Overall, 10.6% of patients in the clopidogrel group and 12.5%of patients in the placebo group experienced one of the primaryoutcomes. This occurred with a relative risk of 0.84 (P = 0.001).For patients who underwent CABG treatment, primary outcomesoccurred in 16.2% of the placebo group versus 14.5% in theclopidogrel group. Benefits were seen mainly in those patientswho had received combination therapy before the procedure. Forpatients undergoing CABG, there was no significant trend of life-threatening bleeding with clopidogrel, the use of which was con-fined to within 5 days of CABG surgery. According to the CUREtrial, the benefits of starting clopidogrel with aspirin in non-STMI outweighed the hemorrhagic risk, even in patients treatedby CABG. Clopidogrel-treated patients also experienced reducedin-hospital refractory ischemia, recurrent angina, and heart fail-ure. The CURE trial concluded that clopidogrel is beneficialin ACS patients whether or not they undergo revascularization.With regard to maximizing the clinical benefit and minimizingthe hemorrhagic risk associated with clopidogrel and CABG, theresults of the CURE trial suggested initiating dual therapy uponpresentation and stopping clopidogrel 5 days before the CABGprocedure.

212 Mousa

The PCI-CURE (Percutaneous Coronary Intervention-Clopidogrel in Unstable angina to Prevent Recurrent Events) trial(42), published in 2001, was a sub-analysis of the CURE trialin which 2,658 patients who were included in the CURE studyand had undergone PCI were studied. Primary outcomes werecardiovascular death, MI, or urgent revascularization within 30days of the intervention. PCI patients were given clopidogrel orplacebo in combination with aspirin for 4 weeks prior to surgery.They resumed drug treatment post-surgically and were assessedfor long-term effects up to 1 year later.

There was a significant difference in cardiovascular death andMI between the two groups: 12.6% in the placebo group and8.8% in the clopidogrel group experienced primary outcomes(P = 0.002). PCI-CURE investigators concluded that comparedto placebo, patients with non-ST elevation MI treated with clopi-dogrel plus aspirin and PCI had a reduced risk of cardiovasculardeath and MI by about a third. Long-term therapy was also asso-ciated with a lower rate of cardiovascular death, MI, or the needfor revascularization (P = 0.03). There was no significant differ-ence in major, but not life-threatening, bleeding with clopidogrel.

The results of the PCI-CURE study supported the use ofcombination antiplatelet therapy with clopidogrel plus aspirin inpatients undergoing PCI with ACS and non-ST-segment eleva-tion MI. It was further suggested that long-term combinationtherapy is beneficial in reducing major cardiovascular events.

2.2.6. The MATCH Study In the MATCH trial, 7,599 high-risk patients with cerebrovas-cular disease were treated with clopidogrel or clopidogrel plusaspirin and assessed for vascular events (43). Patients admittedinto this trial were stable and receiving clopidogrel (75 mg/day),and were randomized to either placebo or low-dose aspirin(75 mg/day). Patients had to have experienced an ischemic strokeor transient ischemic attack (TIA) within the past 3 months, andthey had to have one or more of the following four risk fac-tors: previous MI, angina pectoris, diabetes mellitus, and symp-toms of PAD within the past 3 years. The primary end pointfor the MATCH trial was a composite of ischemic stroke, MI,vascular death, and rehospitalization for an acute ischemic event(angina pectoris, worsening PAD symptoms, or TIA). Secondaryoutcomes included death from any type of stroke.

The MATCH trial failed to show any significant differencein risk reduction between the clopidogrel plus aspirin group ver-sus the clopidogrel plus placebo group. There was no statisticalsignificance in any of the primary end points between the twoarms that indicated a reduction in rates of ischemic stroke, MI,vascular death, or rehospitalization for any acute ischemic event.In addition, there was a significantly higher rate of major bleed-ing in the combination therapy group (P< 0.0001). The risk was


1.36% higher in the combination arm as compared to the clopido-grel monotherapy arm. Thus, there was no significant reductionin vascular events by adding aspirin to clopidogrel for high-riskpatients with ischemic stroke or TIA.

2.2.7. The CARESSStudy

In the CARESS (Clopidogrel and Aspirin for Reduction ofEmboli in Symptomatic Carotid Stenosis) trial, Markus et al. (44)evaluated combination therapy in cerebrovascular risk patients.All patients were recently symptomatic, with ≥50% carotid steno-sis with micro-embolic signals (MES), as detected by transcra-nial Doppler ultrasound. Patients were randomized to clopido-grel (loading dose of 300 mg on day 1 followed by 75 mg/day)or placebo. All patients also received aspirin 75 mg/day for theduration of the study. Patients were all >18 years of age, had >50%carotid stenosis, and had experienced ipsilateral carotid territoryTIA or stroke within the preceding 3 months. The concomitantuse of anticoagulants, thrombolytic agents, analgesics (other thanacetaminophen or opioids), and any additional antiplatelet agentswas prohibited during the course of the study. The CARESS trialassessed the proportion of patients who were MES-positive on a1-h recording conducted on day 7 of the trial. A reduction in theMES value would correspond to decreased markers of platelet andthrombus emboli in the ipsilateral middle cerebral artery, lead-ing to a lower risk of recurrent strokes in this patient population.Safety end points were recorded as any adverse or cerebrovas-cular events, such as TIA, ischemic stroke, or cerebral hemor-rhage. Bleeding events were divided into three categories: life-threatening, major, or minor bleeding.

Of the 107 randomized patients who had an MES value >1,43.8% of patients in the combination group versus 72.7% in theplacebo group had a positive MES reading on day 7, favoringcombination therapy. There was a relative risk reduction of 39.8%for those patients who received clopidogrel plus aspirin versusaspirin alone. This was statistically significant (P = 0.0046), withno further increases in bleeding events.

The CARESS study concluded that in “actively embolizing”patients with recently symptomatic carotid stenosis (>50%), thecombination of clopidogrel plus aspirin therapy is more effec-tive than aspirin alone in reducing asymptomatic embolization,with a relative risk reduction of 40% in 7 days. Active treatmentresponse was seen with a similar magnitude of effect on day 2.These findings were in contrast to the results of the MATCHstudy; however, there are some differences to be noted betweenthe two patient populations. The MATCH study data were basedon all types of ischemic stroke, including small-vessel disease,which has the lowest risk of early recurrent stroke because it isnot a process caused by embolism from an atherosclerotic plaque(45). This may be one of the reasons why interpretation of the

214 Mousa

MATCH trial data at 18 months is not consistent with that ofthe CARESS trial after 7 days. The MATCH trial also includedpatients who were experiencing events several weeks after theiracute phase. This group of patients is at the highest risk for recur-rent stroke. In contrast, the CARESS study was consistent withreducing recurrent stroke in patients with large-vessel atheroscle-rotic stroke in the acute phase. The difference in the power ofthese two studies should also be noted. The CARESS study ana-lyzed 107 patients versus 7,276 patients analyzed in the MATCHtrial. Upon direct comparison, the MATCH trial had a greaterpower value than the CARESS study, which could also influencethe observed outcomes.

2.2.8. Summary:Combination Aspirin andClopidogrel Therapy

There is considerable clinical evidence supporting the use of com-bination antiplatelet therapy with clopidogrel and aspirin. TheCLARITY-TIMI and the COMMIT trials both studied patientssuffering from acute MI with ST-segment elevations or newlydeveloped left-bundle branch block (39, 40). These two trialsdemonstrated greater efficacy when patients were given clopi-dogrel plus aspirin versus aspirin alone. Death, reinfarction, andstroke were all reduced with the combination of antiplatelets,with no significant difference in major bleeding compared toaspirin alone. The CURE trial was conducted to investigatethe reduction of risks in patients undergoing vascular interven-tions. Again, patients who received clopidogrel and aspirin beforesurgery experienced a significant reduction in the primary endpoints. These results were supported by the PCI-CURE study(42). Asymptomatic patients who were considered to be at highrisk for atherothrombotic events did not benefit from combina-tion antiplatelet therapy in the CHARISMA study (41). Clopi-dogrel plus aspirin is not recommended for the prevention ofatherothrombotic events in stable patients who have multiple riskfactors alone.

The MATCH and CARESS trials raised much controversydue to their conflicting results. The CARESS trial observedpatients who experienced ipsilateral carotid territory TIA/strokeonly (43–45). Furthermore, observations in the MATCH trialwere recorded up to 18 months, whereas the CARESS resultswere recorded after 7 days.

One issue that deserves further discussion is the durationof therapy. There is conflicting evidence from the MATCH andCARESS trials as to the optimal duration of antiplatelet ther-apy in cerebrovascular disease. For coronary artery disease, theCHARISMA trial failed to show a benefit of long-term clopido-grel in the overall trial population, although the 80% of patientswith clinically evident atherothrombosis experienced a modestreduction of the primary endpoint, and emerging data withdrug-eluting stents suggest that dual antiplatelet therapy may


be required even beyond 1 year. Clearly, additional studies areneeded to evaluate optimal antiplatelet therapy combinations andduration of therapies to permit maximal benefit with the mini-mum of harm in patients with cardiovascular disease.

3. Expert Opinionand FutureDirections

Platelet aggregation plays a key role in the pathogenesis ofcoronary thrombosis, and pharmacologic inhibition of plateletfunction forms the cornerstone of treatment for ACS. Cardio-vascular treatment with clopidogrel versus aspirin has been shownto be beneficial, but with adequate precautions (46). When clopi-dogrel and aspirin were each studied in monotherapy, clopido-grel proved to be superior to aspirin in reducing the risk ofischemic events (47). When the two were combined in the firsttrial to show effectiveness versus aspirin monotherapy, the com-bination was associated with a high risk of bleeding (46). Onlywhen investigators began to observe carefully specific patient pop-ulations did combination therapy begin to demonstrate promis-ing results (e.g., the PCI-CURE study). Clopidogrel has beenshown to be beneficial in the initial treatment and secondary pre-vention of ACS. Compared with aspirin alone, dual antiplatelettherapy with aspirin plus clopidogrel reduces the risk of ves-sel thrombosis and recurrent ischemic events in patients under-going PCI and is a useful adjunct to coronary artery stenting.In CABG surgery, combination aspirin and clopidogrel ther-apy initiated immediately postoperatively improves bypass graftpatency.

The search for ADP receptor antagonists with a rapid onsetand short half-life is ongoing. Such agents would allow for urgentsurgical procedures and would overcome the current limitationsof clopidogrel. Furthermore, antiplatelet agents that work viacollagen receptors, serotonin receptors, or thrombin receptorsmay have additional value and the ability to complement currentantiplatelet therapy. Several rapid-onset and rapid-offset reversibleADP antagonists are currently in clinical development (35, 48).AZD-6140 is a reversible oral P2Y12 receptor antagonist that hasbeen studied in ACS patients in comparison to clopidogrel in theDISPERSE-2 (Dose Confirmation Study Assessing AntiplateletEffects of AZD6140 versus Clopidogrel in Non-ST-Segment Ele-vation MI) study. AZD-6140 exhibited greater mean inhibitionof platelet aggregation than a standard regimen of clopidogrel inACS patients. In addition, AZD-6140 further suppressed plateletaggregation in clopidogrel-pretreated patients (48).

216 Mousa

One could extrapolate that reversible ADP receptorantagonists would have greater clinical benefits based on ex vivostudies showing that their ability to inhibit platelet aggregation issuperior to other drugs such as clopidogrel. However, as demon-strated by the case of oral GPIIb/IIIa inhibitors, promising labo-ratory results do not necessarily translate into successful outcomesin the clinical setting. Despite consistent evidence of substantialinhibition of platelet aggregation by oral GPIIb/IIIa inhibitors,this class of drugs provided no clinical benefit in phase-III trialsand, in fact, was harmful. The connection between ex vivo inhibi-tion of platelet aggregation and clinical benefit of platelet P2Y12antagonists is also not straightforward. A link between inflamma-tory status and clinical benefit from antiplatelet agents continuesto emerge and highlights the fact that biomarkers beyond ex vivoplatelet aggregation might better predict the clinical benefit ofantiplatelet agents that reduce platelet activation. Ongoing trialshold promise for determining the appropriate targets for maxi-mizing antiplatelet efficacy, but the current lack of a proven exvivo assay that correlates with clinical outcomes hampers clinicalinvestigation, particularly in light of the expense involved in con-ducting these mega trials.

ADP receptor blockers require 3–7 days to reach maximuminhibition of platelet aggregation. When investigators began toassess the use of a loading dose with clopidogrel, therapeuticresults occurred more rapidly (33, 44). There was a rapid onsetof platelet aggregate inhibition, with an antithrombotic effectobserved within 90 min. The optimal type of antiplatelet ther-apy for patients who will undergo surgery would be one with arapid onset and a relatively short elimination half-life, allowingfor a once- or twice-daily regimen.

A significant impediment to the development and clinicalapplication of antiplatelet therapies is the prohibitive cost of clin-ical trials and the potential risk involved in attaining clinical supe-riority and safety over existing regimens. The recruitment andstudy of a large number of patients for a prolonged period of timemay be necessary to demonstrate a modest clinical benefit. Thus,the prohibitive cost of clinical trials demonstrating significant dif-ferences to current antiplatelet therapies might limit progress inadvancing new antiplatelet targets.

References

1. Jamieson, D.G., Parekh, A., and Ezekowitz,M.D. (2005) Review of antiplatelet ther-apy in secondary prevention of cerebrovas-cular events: a need for direct compar-

isons between antiplatelet agents J Cardio-vasc Pharmacol Ther 10, 153–61.

2. Collaboration, A.T. (1994) Collaborativeoverview of randomised trials of antiplatelet


therapy – I: prevention of death, myocar-dial infarction, and stroke by prolongedantiplatelet therapy in various categories ofpatients. BMJ 308, 81–106.

3. Fitzgerald, D.J., Roy, L., Catella, F., andFitzGerald, G.A. (1986) Platelet activation inunstable coronary disease N Eng J Med 315,983–9.

4. Fuster, V. (1981) Role of platelets in thedevelopment of atherosclerotic disease andpossible interference with platelet inhibitordrugs Scand J Haematol Suppl 38, 1–38.

5. Hamm, C.W., Lorenz, R.L., Bleifeld, W.,Kupper, W., Wober, W., and Weber, P.C.(1987) Biochemical evidence of platelet acti-vation in patients with persistent unstableangina J Am Coll Cardiol 10, 998–06.

6. Mousa, S., and Bennett, J.S. (1996) Plateletsin health and disease: platelet GPIIb/IIIastructure and function; recent advancesin antiplatelet therapy Drugs Future 21,1141–54.

7. Mousa, S., and Guigliano, R. Antiplatelettherapies: platelet GPIIb/IIIa antagonistsand beyond. In: Sasahara, A., ed. New Ther-apeutic Agents in Thrombosis. New York:Marcel Dekker;2003:341–7.

8. Born, G.V., Gorog, P., and Begent, N.A.(1983) The biologic background to sometherapeutic uses of aspirin Am J Med 74, 2–9.

9. Mousa, S.A. (2003) Antiplatelet therapies:platelet GPIIb/IIIa antagonists and beyondCurr Pharm Des 9, 2317–22.

10. Hart, R.G., and Harrison, M.J. (1996)Aspirin wars: the optimal dose of aspirin pre-vent stroke Stroke 27, 585–7.

11. Hennekens, C.H., Sechenova, O., Hollar,D., and Serebruany, V.L. (2006) Dose ofaspirin in the treatment and prevention ofcardiovascular disease: current and futuredirections J Cardiovasc Pharmacol Ther 11,170–6.

12. Haynes, R.B., Sandler, R.S., Larson, E.B.,Pater, J.L., and Yatsu, F.M. (1992) A criti-cal appraisal of ticlopidine, a new antiplateletagent. Effectiveness and clinical indicationsfor prophylaxis of atherosclerotic events ArchIntern Med 152, 1376–80.

13. Love, B.B., Biller, J., and Gent, M. (1998)Adverse haematological effects of ticlopi-dine. Prevention, recognition and manage-ment Drug Saf 19, 89–98.

14. Creager, M.A. (1998) Results of theCAPRIE trial: efficacy and safety of clopido-grel. Clopidogrel versus aspirin in patients atrisk of ischaemic events Vasc Med 3, 257–60.

15. Moussa, I., Oetgen, M., Roubin, G.,Colombo, A., Wang, X., Iyer, S., Maida,R., Collins, M., Kreps, E., and Moses,

J.W. (1999) Effectiveness of clopidogrel andaspirin versus ticlopidine and aspirin in pre-venting stent thrombosis after coronary stentimplantation Circulation 99, 2364–6.

16. Hankey, G.J., Sudlow, C.L., and Dun-babin, D.W. (2000) Thienopyridine deriva-tives (ticlopidine, clopidogrel) versus aspirinfor preventing stroke and other serious vas-cular events in high vascular risk patientsCochrane Database Syst Rev CD001246.

17. Bednar, M.M., Quilley, J., Russell, S.R.,Fuller, S.P., Booth, C., Howard, D., andGross, C.E. (1996) The effect of oralantiplatelet agents on tissue plasminogenactivator-mediated thrombolysis in a rab-bit model of thromboembolic stroke Neuro-surgery 39, 352–9.

18. Nicolau, D.P., Tessier, P.R., and Nightingale,C.H. (1999) Beneficial effect of combina-tion antiplatelet therapy on the developmentof experimental Staphylococcus aureus endo-carditis Int J Antimicrob Agents 11, 159–61.

19. Lenz, T.L., and Hilleman, D.E. (2000)Aggrenox: a fixed-dose combination ofaspirin and dipyridamole Ann Pharmacother34, 1283–90.

20. Steinhubl, S.R., Berger, P.B., Mann, J.T.,3rd, Fry, E.T., DeLago, A., Wilmer, C., andTopol, E.J. (2002) Early and sustained dualoral antiplatelet therapy following percuta-neous coronary intervention: a randomizedcontrolled trial JAMA 288, 2411–20.

21. Bertrand, M.E., Rupprecht, H.J., Urban, P.,and Gershlick, A.H. (2000) Double-blindstudy of the safety of clopidogrel with andwithout a loading dose in combination withaspirin compared with ticlopidine in combi-nation with aspirin after coronary stenting: the clopidogrel aspirin stent internationalcooperative study (CLASSICS) Circulation102, 624–9.

22. Mousa, S.A. (1999) Antiplatelet therapies:recent advances in the development ofplatelet glycoprotein IIb/IIIa antagonistsCurr Interv Cardiol Rep 1, 243–52.

23. Mousa, S.A. (1999) Antiplatelet therapies:from aspirin to GPIIb/IIIa-receptor antag-onists and beyond Drug Discov Today 4,552–61.

24. Phillips, D.R., Charo, I.F., and Scarborough,R.M. (1991) GPIIb-IIIa: the responsive inte-grin Cell 65, 359–62.

25. Pytela, R., Pierschbacher, M.D., Ginsberg,M.H., Plow, E.F., and Ruoslahti, E. (1986)Platelet membrane glycoprotein IIb/IIIa:member of a family of Arg-Gly-Asp – specificadhesion receptors Science 231, 1559–62.

26. The EPIC Investigators (1994) Use of amonoclonal antibody directed against the

218 Mousa

platelet glycoprotein IIb/IIIa receptor inhigh-risk coronary angioplasty. N Engl J Med330, 956–61.

27. Fitchett, D.H., Langer, A., Armstrong, P.W.,Tan, M., Mendelsohn, A., and Goodman,S.G. (2006) Randomized evaluation of theefficacy of enoxaparin versus unfractionatedheparin in high-risk patients with non-ST-segment elevation acute coronary syndromesreceiving the glycoprotein IIb/IIIa inhibitoreptifibatide. Long-term results of the Inte-grilin and Enoxaparin Randomized Assess-ment of Acute Coronary Syndrome Treat-ment (INTERACT) trial Am Heart J 151,373–9.

28. Hayes, R., Chesebro, J.H., Fuster, V., Dan-gas, G., Fallon, J.T., Sharma, S.K., Coller,B.S., Badimon, L., Marmur, J.D., and Badi-mon, J.J. (2000) Antithrombotic effects ofabciximab Am J Cardiol 85, 1167–72.

29. Januzzi, J.L., Jr., Snapinn, S.M., DiBattiste,P.M., Jang, I.K., and Theroux, P. (2002)Benefits and safety of tirofiban among acutecoronary syndrome patients with mild tomoderate renal insufficiency: results fromthe Platelet Receptor Inhibition in IschemicSyndrome Management in Patients Limitedby Unstable Signs and Symptoms (PRISM-PLUS) trial Circulation 105, 2361–6.

30. Lincoff, A.M., Califf, R.M., Van de Werf,F., Willerson, J.T., White, H.D., Armstrong,P.W., Guetta, V., Gibler, W.B., Hochman,J.S., Bode, C., Vahanian, A., Steg, P.G.,Ardissino, D., Savonitto, S., Bar, F., Sad-owski, Z., Betriu, A., Booth, J.E., Wolski, K.,Waller, M., and Topol, E.J. (2002) Mortal-ity at 1 year with combination platelet gly-coprotein IIb/IIIa inhibition and reduced-dose fibrinolytic therapy vs conventional\hbox{fibrinolytic} therapy for acute myocar-dial infarction: GUSTO V randomized trialJAMA 288, 2130–5.

31. Neumann, F.J., Kastrati, A., Schmitt, C.,Blasini, R., Hadamitzky, M., Mehilli, J.,Gawaz, M., Schleef, M., Seyfarth, M.,Dirschinger, J., and Schomig, A. (2000)Effect of glycoprotein IIb/IIIa receptorblockade with abciximab on clinical andangiographic restenosis rate after the place-ment of coronary stents following acutemyocardial infarction J Am Coll Cardiol 35,915–21.

32. Anderson, J.L., Adams, C.D., Antman, E.M.,Bridges, C.R., Califf, R.M., Casey, D.E.,Jr., Chavey, W.E., 2nd, Fesmire, F.M.,Hochman, J.S., Levin, T.N., Lincoff, A.M.,Peterson, E.D., Theroux, P., Wenger, N.K.,Wright, R.S., Smith, S.C., Jr., Jacobs, A.K.,Halperin, J.L., Hunt, S.A., Krumholz, H.M.,

Kushner, F.G., Lytle, B.W., Nishimura, R.,Ornato, J.P., Page, R.L., and Riegel, B.(2007) ACC/AHA 2007 guidelines forthe management of patients with unstableangina/non-ST-elevation myocardial infarc-tion: a report of the American College of Car-diology/American Heart Association TaskForce on Practice Guidelines (Writing Com-mittee to Revise the 2002 Guidelines forthe Management of Patients With Unsta-ble Angina/Non-ST-Elevation MyocardialInfarction) developed in collaboration withthe American College of Emergency Physi-cians, the Society for Cardiovascular Angiog-raphy and Interventions, and the Society ofThoracic Surgeons endorsed by the AmericanAssociation of Cardiovascular and PulmonaryRehabilitation and the Society for AcademicEmergency Medicine J Am Coll Cardiol 50,e1–157.

33. Cadroy, Y., Bossavy, J.P., Thalamas, C.,Sagnard, L., Sakariassen, K., and Boneu,B. (2000) Early potent antithromboticeffect with combined aspirin and a loadingdose of clopidogrel on experimental arterialthrombogenesis in humans Circulation 101,2823–8.

34. Coma-Canella, I., Velasco, A., and Castano,S. (2005) Prevalence of aspirin resistancemeasured by PFA-100 Int J Cardiol 101,71–6.

35. Wiviott, S.D., Braunwald, E., McCabe,C.H., Montalescot, G., Ruzyllo, W., Got-tlieb, S., Neumann, F.J., Ardissino, D., DeServi, S., Murphy, S.A., Riesmeyer, J., Weer-akkody, G., Gibson, C.M., and Antman,E.M. (2007) Prasugrel versus clopidogrel inpatients with acute coronary syndromes NEng J Med 357, 2001–15.

36. Angiolillo, D.J., Shoemaker, S.B., Desai, B.,Yuan, H., Charlton, R.K., Bernardo, E.,Zenni, M.M., Guzman, L.A., Bass, T.A., andCosta, M.A. (2007) Randomized compari-son of a high clopidogrel maintenance dosein patients with diabetes mellitus and coro-nary artery disease: results of the Optimiz-ing Antiplatelet Therapy in Diabetes Mel-litus (OPTIMUS) study Circulation 115,708–16.

37. Spertus, J.A., Kettelkamp, R., Vance, C.,Decker, C., Jones, P.G., Rumsfeld, J.S.,Messenger, J.C., Khanal, S., Peterson,E.D., Bach, R.G., Krumholz, H.M., andCohen, D.J. (2006) Prevalence, predic-tors, and outcomes of premature discon-tinuation of thienopyridine therapy afterdrug-eluting stent placement: results fromthe PREMIER registry Circulation 113,2803–9.


38. Smith, S.C., Jr., Feldman, T.E., Hirshfeld,J.W., Jr., Jacobs, A.K., Kern, M.J., King,S.B., 3rd, Morrison, D.A., O′Neill, W.W.,Schaff, H.V., Whitlow, P.L., Williams, D.O.,Antman, E.M., Adams, C.D., Anderson,J.L., Faxon, D.P., Fuster, V., Halperin, J.L.,Hiratzka, L.F., Hunt, S.A., Nishimura, R.,Ornato, J.P., Page, R.L., and Riegel, B.(2006) ACC/AHA/SCAI 2005 guidelineupdate for percutaneous coronary interven-tion: a report of the American Collegeof Cardiology/American Heart Associa-tion Task Force on Practice Guidelines(ACC/AHA/SCAI Writing Committee toUpdate the 2001 Guidelines for Percuta-neous Coronary Intervention) J Am CollCardiol 47, e1–121.

39. Sabatine, M.S., Cannon, C.P., Gibson, C.M.,Lopez-Sendon, J.L., Montalescot, G., Ther-oux, P., Claeys, M.J., Cools, F., Hill, K.A.,Skene, A.M., McCabe, C.H., and Braunwald,E. (2005) Addition of clopidogrel to aspirinand fibrinolytic therapy for myocardial infarc-tion with ST-segment elevation N Eng J Med352, 1179–89.

40. Chen, Z.M., Jiang, L.X., Chen, Y.P., Xie,J.X., Pan, H.C., Peto, R., Collins, R., andLiu, L.S. (2005) Addition of clopidogrelto aspirin in 45,852 patients with acutemyocardial infarction: randomised placebo-controlled trial Lancet 366, 1607–21.

41. Bhatt, D.L., Fox, K.A., Hacke, W., Berger,P.B., Black, H.R., Boden, W.E., Cacoub, P.,Cohen, E.A., Creager, M.A., Easton, J.D.,Flather, M.D., Haffner, S.M., Hamm, C.W.,Hankey, G.J., Johnston, S.C., Mak, K.H.,Mas, J.L., Montalescot, G., Pearson, T.A.,Steg, P.G., Steinhubl, S.R., Weber, M.A.,Brennan, D.M., Fabry-Ribaudo, L., Booth,J., and Topol, E.J. (2006) Clopidogrel andaspirin versus aspirin alone for the preven-tion of atherothrombotic events N Eng J Med354, 1706–17.

42. Mehta, S.R., Yusuf, S., Peters, R.J., Bertrand,M.E., Lewis, B.S., Natarajan, M.K., Malm-berg, K., Rupprecht, H., Zhao, F., Chrolavi-

cius, S., Copland, I., and Fox, K.A. (2001)Effects of pretreatment with clopidogrel andaspirin followed by long-term therapy inpatients undergoing percutaneous coronaryintervention: the PCI-CURE study Lancet358, 527–33.

43. Diener, H.C., Bogousslavsky, J., Brass, L.M.,Cimminiello, C., Csiba, L., Kaste, M.,Leys, D., Matias-Guiu, J., and Rupprecht,H.J. (2004) Aspirin and clopidogrel com-pared with clopidogrel alone after recentischaemic stroke or transient ischaemic attackin high-risk patients (MATCH): randomised,double-blind, placebo-controlled trial Lancet364, 331–7.

44. Markus, H.S., Droste, D.W., Kaps, M., Lar-rue, V., Lees, K.R., Siebler, M., and Ringel-stein, E.B. (2005) Dual antiplatelet ther-apy with clopidogrel and aspirin in symp-tomatic carotid stenosis evaluated usingdoppler embolic signal detection: the Clopi-dogrel and Aspirin for Reduction of Emboliin Symptomatic Carotid Stenosis (CARESS)trial Circulation 111, 2233–40.

45. Rothwell, P.M. (2004) Lessons fromMATCH for future randomised trials in sec-ondary prevention of stroke Lancet 364,305–7.

46. Hirsh, J., and Bhatt, D.L. (2004) Compar-ative benefits of clopidogrel and aspirin inhigh-risk patient populations: lessons fromthe CAPRIE and CURE studies Arch InternMed 164, 2106–10.

47. CAPRIE Steering Committee (1996) A ran-domised, blinded, trial of clopidogrel versusaspirin in patients at risk of ischaemic events(CAPRIE). Lancet 348, 1329–39.

48. Storey, R.F., Husted, S., Harrington, R.A.,Heptinstall, S., Wilcox, R.G., Peters, G.,Wickens, M., Emanuelsson, H., Gurbel, P.,Grande, P., and Cannon, C.P. (2007) Inhi-bition of platelet aggregation by AZD6140,a reversible oral P2Y12 receptor antagonist,compared with clopidogrel in patients withacute coronary syndromes J Am Coll Cardiol50, 1852–6.

Chapter 8

Prasugrel: A Novel Platelet ADP P2Y12 Receptor Antagonist

Shaker A. Mousa, Walter P. Jeske, and Jawed Fareed

Abstract

Novel adenosine diphosphate (ADP) P2Y12 antagonists such as prasugrel, ticagrelor, cangrelor, andelinogrel are in various phases of clinical development. These ADP P2Y12 antagonists have advantagesover clopidogrel ranging from faster onset to greater and less variable inhibition of platelet function.Novel ADP P2Y12 antagonists are under investigation to determine whether their use can result inimproved antiplatelet activity, faster onset of action, and/or greater antithrombotic effects than clopi-dogrel without an unacceptable increase in hemorrhagic or other side effects. Prasugrel (CS-747; LY-640315), a novel third-generation oral thienopyridine, is a specific, irreversible antagonist of the plateletADP P2Y12 receptor. Pre-clinical and early phase clinical studies have shown prasugrel to be characterizedby more potent antiplatelet effects, lower inter-individual variability in platelet response, and faster onsetof activity compared to clopidogrel. Recent findings from large-scale phase-III testing show prasugrelto be more efficacious in preventing ischemic events in acute coronary syndrome patients undergoingpercutaneous coronary intervention (PCI); however, this is achieved at the expense of an increased riskof bleeding. Prasugrel provides more rapid and consistent platelet inhibition than clopidogrel.

Key words: Antiplatelet, acute coronary syndrome, antithrombotic, percutaneous coronary inter-vention, platelets, thienopyridines, thrombosis, antiplatelet combinations.

1. Introduction

Platelets are the principle effectors of cellular hemostasis and keymediators in the pathogenesis of thrombosis. A variety of mem-brane receptors determine platelet reactivity with numerous ago-nists and adhesive proteins, and, therefore, represent key targetsfor the development of antiplatelet drug therapies. In this regard,several rapid-onset and rapid-offset reversible ADP antagonistsare in clinical development, including reversible oral and rapidacting intravenous (1, 2) P2Y12 receptor antagonists (Table 8.1).


221

222 Mousa, Jeske, and Fareed

Table 8.1Promising new antiplatelet P2Y12 inhibitor therapies indevelopment

AgentPotential advantages over current agents(clopidogrel, ticlopidine)

Prasugrel Faster onset, greater antiplatelet effect, less variableresponse (FDA approved)

AZD6140 Reversible, faster onset and offset, greater antiplateleteffect

Cangrelor (IV) Reversible, shorter half-life, faster onset, greaterantiplatelet effect, less variable response

Novel inhibitors of platelet adhesion in early development targetthe von Willebrand (vWF)-GPIb/IX and collagen/GPVI interac-tions. Since platelet aggregation also plays such a critical role inthe pathogenesis of arterial thrombosis, more potent agents thatinterfere with platelet aggregation via other pathways (e.g., thethrombin receptor) are also under clinical investigation (3)

The major limitation to treatment with multiple antiplateletagents is the increased bleeding risk associated with the enhancedantiplatelet effect. This is exemplified by the clinical conun-drum in patients with acute coronary syndrome (ACS) who mayneed to undergo coronary artery bypass graft (CABG) surgery.Aspirin and clopidogrel irreversibly inhibit platelet function, withmaximal antiplatelet effect occurring after 3–5 days of treat-ment. The increased risk of procedural bleeding arising fromdual aspirin and clopidogrel administration immediately prior toCABG surgery raises the question of whether clopidogrel shouldbe routinely given to patients presenting with ACS. In light ofthe current recommendation to discontinue clopidogrel at least5 days prior to elective CABG surgery, the emergency physicianis likely to avoid clopidogrel in anticipation that the patient mayrequire urgent cardiac surgery. Delaying clopidogrel therapy untilcoronary revascularization has been performed would, however,deprive patients of the early clinical benefits of the drug. Theselimitations might be solved with the availability of rapid-onset andrapid-offset ADP antagonists.

Furthermore, it is becoming clear that there is variability inindividual responses to clopidogrel, with reported rates of inad-equate antiplatelet response ranging between 4 and 30% (4).Reasons for this include genetic variables (polymorphisms of theP2Y12 receptor or CYP3A4 pathway), up-regulation of alternativepathways of platelet activation, and greater baseline pre-treatmentplatelet reactivity, as well as extrinsic mechanisms such as patientnon-compliance and drug–drug interactions involving CYP3A4(4). Hence, a clinical need exists for superior antiplatelet agents.

Prasugrel: A Novel Platelet ADP P2Y12 Receptor Antagonist 223

2. Prasugrel

2.1.Pharmacokinetics

Prasugrel is a pro-drug requiring activation by the hepatic CYPsystem. As it is rapidly absorbed, prasugrel has a faster onsetof action compared to clopidogrel, with peak concentrations ofactive metabolite seen at 30 min (5). Prasugrel’s active metaboliteis predominantly renally excreted (approximately 70%) and has amean elimination half-life of 3.7 h. Important metabolic differ-ences between clopidogrel and prasugrel result in a higher con-centration of active metabolite with prasugrel. Approximately 85%of clopidogrel is hydrolyzed by esterases to an inactive carboxylicacid derivative, leaving 15% of the pro-drug to be metabolized ina two-step CYP process into active metabolite. Prasugrel, by con-trast, is rapidly hydrolyzed by carboxyesterases and then metabo-lized in a single, CYP-dependent step that uses primarily isotypesCYP3A4 and CPY2B6 (6), which translates into improved inhi-bition of platelet aggregation on a mg/kg basis compared withclopidogrel.

Current evidence from phase-I studies demonstrates that pra-sugrel, when compared to clopidogrel, provides greater inhibi-tion of platelet aggregation with more rapid onset and less non-responsiveness, most likely due to more efficient generation ofthe prasugrel active metabolite. In one study, 68 healthy subjectsnot taking aspirin received either a 300-mg loading dose of clopi-dogrel (then 75 mg daily) or a 60-mg loading dose of prasugrel(then 10 mg daily), followed by the alternate therapy after a 2-week washout (7). The peak inhibitory effect on platelet aggre-gation was greater with prasugrel (mean inhibition, 79% versus35%, P<0.001), while onset of antiplatelet activity was more rapid(maximal inhibitory effect achieved in 60 min with prasugrel ver-sus 4–6 h with clopidogrel). Drug resistance in this study wasdefined as <20% inhibition of platelet aggregation (IPA) at 24 hand was seen in 42% of the clopidogrel-treated patients and noneof the prasugrel-treated patients (7).

2.2. TherapeuticReview

A multiple oral dose phase-I study involving 30 patients demon-strated benefit in terms of the maximum level of platelet inhibi-tion for all prasugrel doses (5, 10, or 20 mg) over clopidogrel75 mg (P<0.001) (8). A 60-mg loading dose was been shownto provide faster onset of action and greater IPA than either a300- or 600-mg loading dose of clopidogrel (9). Maintenancedoses of 10 and 15 mg daily of prasugrel were superior to clopi-dogrel 75 mg daily in a phase-Ib study, with greater platelet inhi-bition (61 versus 68 versus 30%, P<0.0001) and less incidenceof non-responsiveness (10). Such results supported proceedingto phase-II testing, which has demonstrated an acceptable safety


profile for prasugrel as compared to clopidogrel. In particular, inthe JUMBO-TIMI 26 trial, which compared prasugrel to clopi-dogrel use in 900 patients undergoing elective or urgent percu-taneous coronary intervention (PCI), no statistical difference inTIMI major and minor bleeding was noted (1.7 versus 1.2%, P =0.59). The composite endpoint of 30-day major adverse cardiacevents trended lower in the prasugrel-treated patients (7.2 versus9.4%, P=0.26), but was not statistically significant (11).

The PRINCIPLE-TIMI 44 study was the first to compareprasugrel 10 mg daily (after a 60-mg loading dose) to highdose clopidogrel (150 mg daily after a 600-mg loading dose)in patients undergoing catheterization for planned PCI (12).Prasugrel-treated patients had more consistent levels of plateletinhibition with lower inter-individual variability. IPA at 6 hwas significantly higher in the prasugrel group (74 versus 31%,P<0.0001), and the effect was maintained at 14 days. No TIMImajor bleeds occurred in this study of 201 patients, while TIMIminor bleeds occurred in 2 patients (2%) in the prasugrel groupversus none in the clopidogrel group.

Data from the phase III TRITON-TIMI 38 (Trial to AssessImprovement in Therapeutic Outcomes by Optimizing PlateletInhibition with Prasugrel-TIMI 38) trial demonstrated improvedclinical outcomes with prasugrel as compared to clopidogrel.TRITON-TIMI 38 was a double-blind, randomized controlledtrial comparing prasugrel with clopidogrel in 13,608 patientswith moderate- to high-risk ACS (26% STEMI, 74% unstableangina or NSTEMI) who underwent PCI (13). Patients wererandomized to treatment with either prasugrel (60-mg loadingdose then 10 mg/day maintenance) or clopidogrel (300-mg load-ing dose then 75 mg /day). The median duration of therapywas 14.5 months. The primary endpoint, a composite of cardio-vascular death, nonfatal myocardial infarction (MI) or nonfatalstroke, occurred in 9.9% of the prasugrel group versus 12.1% ofclopidogrel-treated patients (hazard ratio 0.81, 95% confidenceinterval, 0.73–0.90; P<0.001), supporting the superior efficacy ofprasugrel. This benefit persisted throughout the follow-up period,suggesting a continued benefit of greater platelet inhibition dur-ing the maintenance phase of therapy, and was evident in bothsubgroups of unstable angina/NSTEMI and STEMI. The pra-sugrel group also showed a significant reduction in the secondaryendpoints of death from cardiovascular causes, nonfatal infarc-tion, or urgent target vessel revascularization at 30 days (HR0.78, P=0.02) and at 90 days (HR 0.79, P<0.001). Significantreductions in the prasugrel group were seen in the rates of MI(7.3% versus 9.4%, HR 0.76, P<0.001), urgent target vessel revas-cularization (2.5% versus 3.7%, HR 0.66, P<0.001), and stentthrombosis (1.1% versus 2.4%, HR 0.48, P<0.001). Notably, thereductions in stent thrombosis were irrespective of stent type or


timing (early, <30 days and late, >30 days) (14). It should benoted that stent use was non-randomized.

The major drawback for prasugrel identified in this trial wasan increased risk of major bleeding, an important issue given itsassociation with mortality. Major bleeding was observed in 2.4%of the patients receiving prasugrel and 1.8% of those receivingclopidogrel (HR 1.32, 95% 1.03–1.68; P=0.03). The rate of life-threatening bleeding was greater in the prasugrel group (1.4 ver-sus 0.9%, P=0.01), including nonfatal (1.1 versus 0.9%, P=0.01)and fatal bleeding (0.4 versus 0.1%, P=0.002). TIMI minorbleeds were more frequent in the prasugrel group as was majornon-CABG bleeding (14). Despite this increase in bleeding, thepre-specified net clinical benefit analysis, which was defined as thecomposite of efficacy (death from any cause, nonfatal MI, nonfatalstroke) and bleeding endpoints (TIMI major hemorrhage), stillfavored prasugrel (12.2% in the prasugrel group versus 13.9% inthe clopidogrel group; HR 0.87, 95% CI 0.79–0.95: P=0.004).This net clinical benefit for prasugrel existed both early and latein the trial. The use of prasugrel in place of clopidogrel essentiallyprevents 23 acute myocardial infarcts per 1,000 patients treated atthe expense of five additional major bleeds (per 1,000 patients).

Post hoc analysis has identified three subgroups with less netclinical benefit or net harm with the use of prasugrel, driven byexcess bleeding risk: previous cerebrovascular events, age >75 orweight <60 kg (15). Patients with a history of stroke or tran-sient ischemic attack had net harm (HR 1.54, 95% CI 1.02–2.32; P=0.04). This group had no evidence of clinical ben-efit with prasugrel when compared to clopidogrel, as evalu-ated by the primary efficacy endpoint, and had a greater rateof TIMI major bleeding (5.0 versus 2.9%, P=0.06), includingintracranial hemorrhage (2.3 versus 0%, P=0.02). Patients 75years of age or older had no net benefit (HR 0.99, 95% CI0.81−1.21; P=0.92), nor did patients weighing <60 kg (HR1.03, 95% CI 0.69−1.53; P=0.89). Interestingly, two other sub-groups appeared to derive significant net benefit from prasug-rel as compared to clopidogrel: STEMI patients and those withdiabetes.

In the STEMI cohort of 3,500 patients, there was a 21%relative risk reduction in the primary endpoint of cardiovasculardeath, MI, or stroke with prasugrel (10.0 versus 12.4%, HR 0.79,P=0.02) (16, 17). This endpoint reduction was driven by lessrecurrent MI and stent thrombosis with no difference in mor-tality and no increase in major bleeding. A greater reduction inischemic events and MI with no increase in bleeding comparedto clopidogrel was clearly evident in patients with diabetes (30%risk reduction). The reason for this lack of bleeding is unclearbut may represent possible protection from bleeding in STEMIand diabetic patients due to increased platelet activation. Another


phase-III trial in development is the TRILOGY ACS study (Tar-geted Platelet Inhibition to Clarify the Optimal Strategy to Med-ically Manage Acute Coronary Syndromes) (18). This double-blind, randomized controlled trial will evaluate the efficacy andsafety of prasugrel versus clopidogrel in ACS patients who aremedically managed and not planned for revascularization (18).

2.3. Prasugrel inRenal Impairmentand End-Stage RenalDisease

Pharmacokinetic (PK) and pharmacodynamic (PD) responses toprasugrel were compared in three studies of healthy subjects ver-sus those with moderate- or end-stage renal impairment. The datashowed no difference in PK or PD responses between healthysubjects and subjects with moderate renal impairment (19). PKand PD studies with prasugrel in moderate hepatic impairmentalso showed little to no effect on PK or platelet aggregation rela-tive to healthy controls, suggesting that a dose adjustment wouldnot be required in patients with moderate liver diseases (20).

3. Conclusion

Prasugrel is a third-generation thienopyridine that selectivelyinhibits the platelet P2Y12 receptor. It leads to platelet inhibi-tion more rapidly, more potently, and with less inter-individualvariability as compared to clopidogrel. Such pharmacodynamicproperties reflect pharmacokinetic differences, namely a more effi-cient metabolism of prasugrel into its active metabolite comparedto clopidogrel. Phase-III testing in high-risk patients undergo-ing PCI showed long-term prasugrel use translates into improvedclinical outcomes compared to clopidogrel. Increased risk ofTIMI major bleeding was demonstrated but the net clinical out-come still favors prasugrel over clopidogrel. Clearly, minimizingbleeding risk, perhaps by dose modifications in specific popula-tions, will maximize the clinical benefit of prasugrel. Further clin-ical studies will evaluate prasugrel use in other clinical scenariosand hopefully better define the patient populations who will bestbenefit from this novel antiplatelet agent.

The bar has been raised for therapeutic regimens employingaspirin, clopidogrel, and their combination. Antiplatelet therapieswith enhanced efficacy and parallel safety profiles are desired, butsuch agents may be mutually exclusive and difficult to attain.Such was the case in the TRITON-TIMI-38 study (21), inwhich enhanced antiplatelet therapy and a significant reductionin ischemic events with prasugrel were accompanied by a sig-nificant risk of fatal and life-threatening bleeding in compari-son to clopidogrel. Newer antiplatelet agents that target differentmechanisms are also being developed and studied in addition to


aspirin and clopidogrel (22). The major challenges faced by thesenewer agents include not only the difficulty in proving incre-mental meaningful clinical benefit, but doing so with a minimalincrease in bleeding risk against a background of contemporarytherapies which themselves continue to evolve.

References

1. Cannon, C.P., Husted, S., Harrington, R.A.,et al. (2007) Safety, tolerability, and ini-tial efficacy of AZD6140, the first reversibleoral adenosine diphosphate receptor antago-nist, compared with clopidogrel, in patientswith non-ST-segment elevation acute coro-nary syndrome: primary results of theDISPERSE-2 trial J Am Coll Cardiol 50,1844–51.

2. Greenbaum, A.B., Grines, C.L., Bittl, J.A.,et al. (2006) Initial experience with an intra-venous P2Y12 platelet receptor antagonist inpatients undergoing percutaneous coronaryintervention: results from a 2-part, phaseII, multicenter, randomized, placebo- andactive-controlled trial Am Heart J 151, 689.

3. Becker, R.C., Moliterno, D.J., Jennings,L.K., Pieper, K.S., Pei, J., Niederman, A.,Ziada, K.M., Berman, G., Strony, J., Joseph,D., Mahaffey, K.W., Van de Werf, F., Veltri,E., and Harrington, R.A, and for the TRA-PCI Investigators. (2009) Safety and tolera-bility of SCH 530348 in patients undergoingnon-urgent percutaneous coronary interven-tion: a randomized, double-blind, placebo-controlled phase II study Lancet 373,919–28.

4. Nguyen, T.A., Diodati, J.G. and Pharand, C.(2005) Resistance to clopidogrel: a review ofthe evidence JACC 45, 1157–64.

5. Jakubowski, J.A., Payne, C.D., Li, Y.G.,et al. (2008) A comparison of the antiplateleteffects of prasugrel and high-dose clopido-grel as assessed by VASP-phosphorylationand light transmission aggregometry ThrombHaemost 99, 215–22.

6. Farid, N., McIntosh, M., Garofolo, F., et al.(2007) Determination of the active andinactive metabolites of prasugrel in humanplasma by liquid chromatography/tandemmass spectrometry Rapid Commun MassSpectrom 21, 169–79.

7. Brandt, J.T., Payne, C.D., Wiviott, S.D.,et al. (2007) A comparison of prasugreland clopidogrel loading doses on plateletfunction: magnitude of platelet inhibition isrelated to active metabolite formation AmHeart J 153, e9–16.

8. Jakubowski, J.A., Matsushima, N., Asai, F.,et al. (2007) A multiple dose study of pra-sugrel (CS- 747), a novel thienopyridineP2Y12 inhibitor, compared with clopidogrelin healthy humans Br J Clin Pharmacol 63,421–30.

9. Haynes, R.B., Sandler, R.S., Larson, E.B.,et al. (1992) A critical appraisal of ticlo-pidine, a new antiplatelet agent: effective-ness and clinical indications for prophylaxis ofatherosclerotic events Arch Intern Med 152,1376–80.

10. Jernberg, T., Payne, C.D., Winters, K.J.,et al. (2006) Prasugrel achieves greater inhi-bition of platelet aggregation and a lowerrate of non-responders compared with clopi-dogrel in aspirin-treated patients with sta-ble coronary artery disease Eur Heart J 27,1166–73.

11. Wiviott, S.D., Antman, E.M., Winters, K.J.,et al. (2005) Randomized comparison ofprasugrel (CS-747, LY640315), a novelthienopyridine P2Y12 antagonist, with clopi-dogrel in percutaneous coronary interven-tion: results of the Joint Utilization ofMedications to Block Platelets Optimally(JUMBO)-TIMI 26 Trial Circulation 111,3366–73.

12. Wiviott, S.D., Trenk, D., Frelinger, A.L.,et al. (2007) Prasugrel compared with highloading- and maintenance-dose clopidogrelin patients with planned percutaneous coro-nary intervention: the prasugrel in compari-son to clopidogrel for inhibition of plateletactivation and aggregation – thrombolysisin myocardial infarction 44 trial Circulation116, 2923–32.

13. Wiviott, S.D., Braunwald, E., McCabe,C.H., et al. (2007) Prasugrel versusclopidogrel in patients with acute coro-nary syndromes N Engl J Med 357,2001–15.

14. Wiviott, S.D., Braunwald, E., McCabe,C.H., et al. (2008) Intensive oral antiplatelettherapy for reduction of ischaemic eventsincluding stent thrombosis in patients withacute coronary syndromes treated withpercutaneous coronary intervention and


stenting in the TRITONTIMI 38 trial: a sub-analysis of a randomized trial Lancet 371,1353–63.

15. Antman, E.M., Wiviott, S.D., Murphy, S.A.,et al. (2008) Early and late benefits of prasug-rel in patients with acute coronary syndromesundergoing percutaneous coronary interven-tion A TRITON-TIMI 38 (Trial to AssessImprovement in Therapeutic Outcomes byOptimizing Platelet Inhibition with Prasug-rel – Thrombolysis in Myocardial Infarction)J Am Coll Cardiol 51, 2028–33.

16. Floyd, J., and Wolfe, S. (2009) Prasug-rel STEMI subgroup analysis Lancet 373,1845–6.

17. Webster, M.W., and Gladding, P. (2009) Pra-sugrel STEMI subgroup analysis Lancet 373,1846–8.

18. Spinler, S.A., and Rees, C. (2009) Reviewof prasugrel for the secondary prevention ofatherothrombosis J Manag Care Pharm 15,383–95.

19. Small, D.S., Wrishko, R.E., Ernest, C.S. 2nd,Ni, L., Winters, K.J., Farid, N.A., Li, Y.G.,Brandt, J.T., Salazar, D.E., Borel, A.G., Kles,

K.A., and Payne, C.D. (2009) Prasugrelpharmacokinetics and pharmacodynamics insubjects with moderate renal impairment andend-stage renal disease J Clin Pharm Ther34, 585–94.

20. Small, D.S., Farid, N.A., Li, Y.G., Ernest,C.S. 2nd, Winters, K.J., Salazar, D.E., andPayne, C.D. (2009) Pharmacokinetics andpharmacodynamic of prasugrel in subjectswith moderate liver disease J Clin PharmTher 34, 575–83.

21. Montalescot, G., Wiviott, S.D., Braunwald,E., Murphy, S.A., Gibson, C.M., McCabe,C.H., Antman, E.M., for the TRITON-TIMI 38 Investigators. (2009) Prasug-rel compared with clopidogrel in patientsundergoing percutaneous coronary inter-vention for ST-elevation myocardial infarc-tion (TRITON-TIMI 38): double-blind,randomized controlled trial Lancet 373,723–31.

22. Anderluh, M., and Dolenc, M.S. (2002)Thrombin receptor antagonists; recentadvances in PAR-1 antagonist developmentCurr Med Chem 9, 1229–50.

Chapter 9

Antithrombotic Effects of Naturally Derived Productson Coagulation and Platelet Function

Shaker A. Mousa

Abstract

To date, there have been few systematic studies of the antiplatelet and/or anticoagulant effects of natu-ral products. According to the Natural Medicines Comprehensive Database, approximately 180 dietarysupplements have the potential to interact with warfarin, and more than 120 may interact with aspirin,clopidogrel, and dipyridamole. These include anise and dong quai (anticoagulant effects); omega 3-fattyacids in fish oil, ajoene in garlic, ginger, ginko, and vitamin E (antiplatelet properties); fucus (heparin-likeactivity); danshen (antithrombin III-like activity and anticoagulant bioavailability); and St. John’s Wortand American Ginseng (interference with drug metabolism). Other supplements, such as high doses ofvitamin E (vitamin K antagonist activity), alfalfa (high-vitamin K content), and coenzyme Q10 (vita-min K-like activity), may affect blood clotting, which is dependent on vitamin K. Studies are needed tounderstand the role of various dietary supplements in thrombosis and their interactions with standardanticoagulants and antiplatelet drugs.

Key words: Dietary supplements, natural products, anticoagulant, antiplatelet, antithrombotic,herbal–drug interaction, dietary supplement–drug interaction, thrombosis, hemostasis, bleeding,cardiovascular diseases, cerebrovascular diseases.

1. Introduction

Cardiovascular and cerebrovascular diseases are the leading causesof death worldwide. The United States alone has an estimated4 million patients on long-term antithrombotic therapies (antico-agulants with or without antiplatelet drugs), and this number isexpected to increase. Surveys suggest that 15–25% of the generalpopulation uses dietary supplements. Recently, 43% of veteransadministration (VA) ambulatory care patients reported using oneor more dietary supplements, with up to 45% of the supplements


229

230 Mousa

taken having potential interactions with medications. Non-prescription drugs, vitamins, minerals, botanicals, homeopathies,and other complementary and alternative therapies are beingactively promoted, resulting in an increased use of these prod-ucts. Because antithrombotic therapies have narrow therapeuticwindows, they must be monitored so that patients are not put atrisk for thromboembolic or hemorrhagic complications.

Drug interactions with herbal and other dietary supplementsare much more difficult to capture, characterize, and predictbecause these products are not required to undergo FDA reviewor approval prior to marketing. They also do not need to meetthe same quality standards as prescription drugs. Manufacturersare responsible for showing simply that they are safe and effica-cious. Based on the pharmacodynamic and pharmacokinetic prop-erties of commercially available supplements and herbal remedies,the potential for interactions is high. Although practitioners areencouraged to report such interactions to the FDA, publishedcase reports of interactions are limited. Clinical guidance for prac-titioners who prescribe antithrombotic medications and patientswho receive them with respect to the use of dietary supplements islacking. Thus, there appears to be a need for increased awarenessamong practitioners about the potential harm or benefit of dietarysupplements among patients receiving long-term antithrombotictherapies.

Although a variety of dietary supplements may affecthemostasis, very few are absolutely contraindicated in people withbleeding disorders or on antithrombotic therapies. The medicalcommunity would benefit from evaluation and information onthese supplements and their harm or benefit to antithrombotictherapies.

Various substances have been proposed and approved aseffective antithrombotics. These agents have been classified asinhibitors of platelet aggregation, the primary causes of arte-rial thrombosis, or coagulation, which causes primarily venousclots. Platelet aggregation inhibitors are designed to target var-ious platelet activation mechanisms, including the thromboxaneA2 and ADP receptor pathways, thrombin receptors, and glyco-protein (GP)IIb/IIIa receptors. Anticoagulant agents functionprimarily by inhibiting the production of fibrin and fibrin accu-mulation within venous blood. Anticoagulant therapies targetthrombin, which converts fibrinogen to fibrin, either directly orindirectly.

A significant amount of research has focused on the individ-ual effects of antiplatelet and anticoagulant agents in treating andpreventing thrombosis. Although studies have shown synergis-tic relationships between aspirin and heparin in reducing cardio-vascular events (1, 2), little research has been conducted on theeffects of combination therapies between platelet and coagulant

Antithrombotic Effects of Naturally Derived Products on Coagulation and Platelet Function 231

inhibitors and their roles as antithrombotic agents. It has beenshown that there is a synergistic relationship between aspirin andwarfarin in decreasing the risk of embolism among patients withprosthetic heart valves, but the effects of combination therapyin direct relation to thrombosis were not clarified (3). In addi-tion, the effects of combining naturally occurring antiplatelet andanticoagulant agents have yet to be determined. The relationshipbetween cancer and the activation of platelets and coagulationhas been established. Tumor cell-induced platelet aggregation notonly plays an important role in thrombosis among cancer patients,but is also recognized as a significant step in the metastatic cascade(4). Likewise, tumors have been shown to activate coagulation byincreasing plasma fibrinogen and fibrin monomer (5).

2. Flavonoids

2.1. Resveratrol Resveratrol (3,4’,5-trihydroxystilbene), a naturally occurringflavonoid found primarily in grapes, red wine, and peanuts (6, 7),has been shown to have antiplatelet properties. It is thoughtthat resveratrol inhibits platelet aggregation induced by colla-gen, thrombin, and ADP (8). The effects of resveratrol onplatelet aggregation seem to be dependent on concentration, andinhibitory results have been observed in vitro and in vivo (8).Despite evidence of its ability to inhibit thrombin-induced plateletactivity, resveratrol has not been shown to affect thrombin-associated fibrin production or posses any other anticoagulantproperties. Furthermore, resveratrol has yet to be linked to theinhibition of platelet aggregation induced by tumor cells. It has,however, shown promise as an anticancer agent, independent ofits ability to inhibit activated platelets. A recent study showed thatflavonoids were able to trigger apoptosis in human leukemia andbreast cancer cells (9). The idea of using resveratrol as a possibleline of treatment in cancer is still in the developmental stage, andmore research is needed (9).

2.2. Tea Extract A major component in green tea, epigallocatechin-3 gallate(EGCG) is believed to be the active ingredient involved in plateletinhibition in humans (10). EGCG primarily targets thrombin-induced platelet activation, and has been shown to decreasethe concentration of the calcium ionophore A23187 (10, 11).Platelet aggregation induced by collagen, thrombin, ADP, andepinephrine is also inhibited by EGCG (11, 12). Catechinspresent in green tea have been found to significantly inhibit thebinding of fibrinogen to the platelet surface GPIIb/IIIa com-plex in humans by decreasing the levels of cytoplasmic calcium

232 Mousa

(11), and it is assumed that EGCG plays a major role in thisinhibitory activity. Green tea stands as a promising antiplateletagent, more so than an anticoagulant, as it does not affect coag-ulation activities, including prothrombin and thrombin time, inhuman citrated plasma (11, 12). Although green tea does notplay a role in preventing coagulation, it does act as an anticanceragent (13). The anticancer activity of green tea catechins, includ-ing EGCG, that possess the gallate group could be mediated inpart through their ability to suppress the tyrosine kinase activity ofthe platelet-derived growth factor (PDGF)-β receptor (14). Theeffectiveness of green tea in inhibiting platelet aggregation mightalso be dependent on the form of the tea. It has been suggestedthat commercially processed tea, in contrast to unprocessed, driedtea leaves, does not decrease thromboxane levels in rats (15). Fur-thermore, only green tea seems to have an antithrombotic effect,while black tea fails to evoke similar effects (16).

2.3. Genistein Found predominantly in soy products, genistein acts as a tyrosinekinase inhibitor and, according to some indications, an inhibitorof platelet aggregation (17, 18). The primary mechanism of inhi-bition of platelet function by genistein is the reduction of cytoso-lic free calcium concentration (18, 19). Genistein has also beenshown to inhibit collagen- and thromboxane analog U46619-induced platelet aggregation (17, 20). In addition to its ability toinhibit activated platelets, genistein has been shown to induce cellcycle arrest and apoptosis, as well as inhibit the growth of cancercells derived from specific cancers of the head, neck, breast, lung,and prostate in culture (21). Genistein also appears to be an effec-tive anti-angiogenic agent with respect to tumor growth (22), butthere is no evidence that the anticancer activity of this flavonoidincludes inhibition of cancer-induced platelet activation. There isconflicting data on the ability of genistein to inhibit thrombin-induced release of serotonin secretions (18, 20). Thus, while it isclear that genistein has potential antiplatelet uses, evidence of anyanticoagulant activity is lacking.

2.4. Echistatin A protein derived from snake venom, echistatin is a 5,000-Da dis-integrin capable of platelet inhibition (23–25). This disintegrinacts as a competitive inhibitor of platelet αIIbβ3 integrin bindingto fibrinogen (23, 26), and decreases phosphorylation to attenu-ate platelet adhesion (25). There is also evidence that echistatinplays a role in the inhibition of collagen (27). To date, whilethe anticoagulant potential of echistatin has yet to be demon-strated, the ability of echistatin to irreversibly bind to integrinαIIbβ3 implicates this disintegrin in the treatment of tumor-induced angiogenesis and tumor cell metastasis (24). Echistatinmay also play a pivotal role in the disassembly of focal adhesionsin fibronectin-adherent B16-BL6 melanoma cells by reducing the


levels of pp125FAK tyrosine phosphorylation (28). Currently, asolid link between the anticancer activity of echistatin and theinhibition of cancer-induced thrombosis has yet to be demon-strated.

2.5. Fisetin Fisetin is a naturally occurring purified protein kinase C inhibitorthat can inhibit platelet aggregation (29). Platelet activity isprimarily suppressed through complete blockage of thrombin-induced shape changes (29, 30). It has been suggested that themechanism of action of fisetin involves inhibition of thromboxaneformation and thromboxane receptor antagonism (30). Fisetininhibits cathepsin G-induced platelet aggregation as well (29).Fisetin has been shown to inhibit the proliferation of normal andcancer cells and induce apoptosis of human promyeloleukemiacells (31, 32). The anticancer properties of fisetin do not appearto extend to cancer-induced platelet aggregation or coagula-tion. Although current findings support the use of fisetin as anantiplatelet agent, research to date has not focused on the poten-tial anticoagulant properties of this flavonoid.

3. Garlic

3.1. Allicin The sulfuric compound responsible for the distinct odor of gar-lic, allicin, appears to be primarily responsible for the inhibitoryeffect of garlic on platelet aggregation. Derived from the cleav-age of alliin by alliin lyase (33), allicin inhibits platelet activityin vitro without affecting cyclooxygenase, lipoxygenase, throm-boxane, vascular prostacyclin synthase, or cyclic AMP levels (34,35). The exact mechanism by which allicin inhibits platelet aggre-gation is unclear, but may be similar to that of ajoene. Ajoeneis a rearrangement of allicin that inhibits platelet aggregation invitro through inhibition of granule release and fibrinogen binding(36). Ajoene is capable of irreversibly inhibiting platelet aggre-gation induced by arachidonic acid, adrenaline, collagen, ADP,and the calcium ionophore A23187 (37). In addition to theirantiplatelet capabilities, allicin and ajoene have been shown toexhibit antitumor properties. Allicin, but not alliin, inhibits theproliferation of human mammary (MCF-2), endometrial, andcolon (HT-29) cancer cells (38). In addition to inhibiting prolif-eration, ajoene also induces apoptosis in human CD34-negativeleukemia cells (39). Currently, allicin shows potential antiplateletactivity.

3.2. Diallyl Trisulfide Diallyl trisulfide is an important paraffinic polysulfide componentof garlic that exhibits certain reversible antiplatelet functions (40).

234 Mousa

Studies have clearly indicated that this polysufide inhibits plateletaggregation and calcium ion mobilization induced by thrombinin a concentration-dependent manner (41). Diallyl trisulfide alsoappears to inhibit thromboxane A2 synthesis, except in a cell-freesystem (34, 40, 41). No effects of diallyl trisulfide on coagu-lation have been documented; however, evidence of chemopre-ventative effects on tumorogenesis have been reported (42). It isthought that diallyl trisulfide inhibits cancer progression by aug-menting the activation of T cells and enhancing the antitumorfunction of macrophages (43). However, these properties do notseem to effect or prevent cancer-induced platelet aggregation orcoagulation.

3.3. Allyl Disulfide There is very little research on the effects of allyl disulfide onplatelet aggregation. It has been suggested that this polysulfideinhibits thromboxane A2, similar to other sulfides present in garlic(34). However, the lack of data on allyl disulfide suggests that thiscomponent plays a minor, if any, role in the antiplatelet activitiesof garlic.

4. Antioxidants

4.1. Vitamin E Along with its antioxidant properties, vitamin E has multipleantiplatelet properties that appear to be independent of its antiox-idant activities (44, 45). Vitamin E has been shown to decreaseplatelet adhesion to collagen, fibrinogen, and fibronectin, andincrease platelet sensitivity to prostaglandin E1 (46–49). Vita-min E also induces the inhibition of protein kinase C (44, 45).Evidence suggests that vitamin E affects phospholipase A2 andat least one other step in the thromboxane A2 cascade, but thatinhibition of platelet aggregation is not exclusively dependent onthese steps within the cascade (50, 51). The reversible effects ofvitamin E on platelet activity are concentration dependent up tomaximal uptake by the platelet, at which point excess vitamin Ehas no effect on function (46, 52). A synergistic effect of vitaminE and other antiplatelet therapies has been suggested to cause anincrease in bleeding (45). Among the numerous mechanisms ofinhibition of platelet activity by vitamin E, there is no evidenceto suggest that this vitamin has anticoagulant properties. Vita-min E may function as an anticancer agent, as it has been shownto inhibit cell proliferation and monocyte adhesion (53). In fact,an early study proposed that supplemental antioxidants such asvitamin E could impede tumor dissemination caused by plateletaggregation (54). Although it has been suggested that the anti-cancer properties of vitamin E are due to its ability to inhibitplatelet aggregation, solid evidence of the reverse does not exist.


4.2. Selenium A deficiency in the trace element selenium enhances platelet acti-vating factor and increases the risk of atherosclerosis (55, 56).Therapies aimed at increasing the levels of selenium in the bodyresult in the inhibition of platelet aggregation, primarily throughthromboxane A2 (57–59). Increases in selenium result in the inhi-bition of thromboxane A2 synthesis in a concentration-dependentmanner in vitro and in vivo and a concomitant decrease in theamount of thromboxane B2 released (52, 58, 59). Increased lev-els of selenium do not affect the biosynthesis of prostaglandin orplatelet adherence to fibrinogen (48, 57). Some studies have sug-gested that selenium may reduce the risk of cancer when properlyincorporated into the diet (60) and that it may have chemopre-ventative effects in breast cancer (61). The specific mechanismsunderlying the anticancer properties of selenium are as yet unde-fined, as are the anticancer effects on platelet activation and coag-ulation. In addition, there is no evidence to date that seleniumfunctions as an anticoagulant.

4.3. Methylsulfonyl-methane(MSM)

The dietary supplement methylsulfonylmethane (MSM) has mul-tiple potential uses, including treatment of athletic injuries, blad-der disorders, and pain syndrome (62), but studies tend to focusdisproportionately on its affects on arthritis and seasonal allergicrhinitis (SAR) (63, 64). MSM is effective in reducing the symp-toms of SAR with relatively low toxicity (2,600 mg/day for 30days) and few side effects, which promotes the safety of MSM(63). MSM also seems to function similarly to aspirin as a can-cer chemopreventive drug, inducing the differentiation of murineerythroleukemia cells (64). Under differentiation-inducing con-ditions and at concentrations reported in other studies, MSMdid not affect prostaglandin E2 or cyclooxygenase activity (64).Thus, if MSM does posses antiplatelet properties, it most likelyaffects aspects of platelet aggregation other than prostaglandinor cyclooxygenase. There is no current evidence of antiplateletor anticoagulant properties of MSM nor has the role of MSM inhuman nutrition been thoroughly studied (62).

5. Anticoagulants

5.1. Heparin Heparin is a naturally occurring indirect thrombin inhibitor inthe body and has been utilized as an anticoagulant for years. Themechanism of action of heparin consists of increasing the activ-ity of antithrombin III to inhibit thrombin synthesis and ulti-mately the production of fibrin (65). However, some studies sug-gest that in addition to acting on antithrombin III, heparin may

236 Mousa

directly affect thrombin (66). Heparin has been proven to pre-vent arterial thrombosis; however, the necessary dosage to doso greatly increases activated partial thromboplastin time (aPTT)(67). Although its ability to prevent arterial thrombosis suggeststhat heparin possesses antiplatelet properties, studies indicate thatheparin is more efficacious when used in the treatment of venousthrombosis (67). Heparin is also indicated as a potential cancertherapeutic. One study in particular suggested that heparin, com-bined with IL-2/LAK therapy, may be useful in preventing fib-rin coagulation on tumor cells (68). This study also suggesteda possible relationship between the reduction of fibrin coagu-lation initiated by tumor cells and the anticoagulant activity ofheparin.

The anticoagulant response produced by heparin is oftenunpredictable, and a major inducer of thrombus growth, namelyfibrin-bound thrombin, is unaffected by heparin (69). A majordrawback to heparin use is the development of heparin-inducedthrombocytopenia, an immunoglobulin-mediated reaction asso-ciated with an increased risk of thrombotic complications (70).Direct thrombin inhibitors, however, are capable of inhibitingfibrin-bound thrombin and provide an alternative treatment forpatients who have developed heparin-induced thrombocytopenia(69). The undesired effects and limitations of heparin use haveled to the development of several alternative heparin derivatives.

5.2. Hirudin One of the most important naturally occurring thrombininhibitors is hirudin, a salivary extract of Hirudo medicinalis (71).Hirudin acts directly on thrombin by binding to several sites,inhibiting all of its functions, including fibrin-bound thrombin(69, 72). Unlike heparin, hirudin is associated with a predica-ble response and does not cause further thrombotic complica-tions, such as thrombocytopenia (73). Hirudin does not appearto cross-react with heparin or heparin derivatives (72). The onlyconcern with hirudin use is a high risk of bleeding (74). Deriva-tives of hirudin are currently in development and show promiseas potent inhibitors of fibrin deposition on clot surfaces, plateletdeposition, and thrombus formation (75). Polyethylene glycol–hirudin may also prove to be an effective agent in treating arte-rial thrombosis (76). The primary use of hirudin currently isin venous thrombosis; it has yet to be evaluated for the treat-ment of arterial thrombosis. Hirudin diminishes the metastaticpotential of tumor cells in the presence and absence of fibrino-gen, suggesting the potential of direct thrombin inhibitors toserve as anticancer agents. Hirudin was able to prevent plateletaggregation induced by human chlorangiocarcinoma (CCA) cellsvia thrombin, suggesting that it may prove useful in prevent-ing tumor cell-induced platelet aggregation or metastasis inCCA (4).


6. Herbal–DrugInteractions:Interactions ofNatural ProductswithAntithromboticDrugs

The impact of dietary supplements on normal hemostasis andantithrombotic therapy should be given careful consideration.Supplements that have been reported to affect normal coagu-lation and platelet activity and/or have been reported to pos-sibly interact with coumarin anticoagulants include danshen(Salvia miltiorrhiza Bunge, Lamiaceae), garlic (Allium sativumL., Lilliaceae), ginkgo (Ginkgo biloba L., Ginkgoaceae), Ameri-can ginseng (Panax quinquefolius L., Araliaceae), Asian ginseng(Panax ginseng C.A. Meyer), and St. John’s wort (Hypericumperforatum L., Clusiaceae). However, most of these reports areeither theoretical or consist of individual cases. In addition tosupplement heterogeneity, individual patient responses reflectingspecific genetic polymorphisms in the cytochrome P450 enzymesystem and other metabolic pathways can alter the metabolismof warfarin (a synthetic dicoumarol, often sold under the tradename Coumadin R©) and/or dietary supplements. Given the vari-ety of dietary supplements available, an important aspect of themanagement of patients on oral anticoagulant therapy is regularassessment of supplement use to ensure that patients are aware ofthe potential risks and benefits of taking supplements in conjunc-tion with their prescribed medications.

References

1. Gaede, A., and Terres, W. (1999) Ther-apy of acute coronary syndrome. Aspirin,heparin, low-molecular-weight heparin,hirudin and GP-IIb/IIIa blockers Herz 24,353–62.

2. Vasudev, S.C., Chandy, T., Sharma, C.P.,Mohanty, M., and Umasankar, P.R. (2000)Synergistic effect of released aspirin/heparinfor preventing bovine pericardial calcificationArtif Organs 24, 129–36.

3. Massel, D., and Little, S.H. (2001) Risksand benefits of adding anti-platelet therapy towarfarin among patients with prosthetic heartvalves: a meta-analysis J Am Coll Cardiol 37,569–78.

4. Akarasereenont, P., Chotewuttakorn, S.,Aiamsa-Ard, T., and Thaworn, A. (2001)The activation of platelet aggregation byhuman cholangiocarcinoma cells is mediatedthrough thrombin receptor J Med Assoc Thai84(Suppl 3), S710–21.

5. Biggerstaff, J.P., Seth, N.B., Meyer, T.V.,Amirkhosravi, A., and Francis, J.L. (1998)Fibrin monomer increases platelet adherence

to tumor cells in a flowing system: a possiblerole in metastasis? Thromb Res 92, S53–8.

6. Bhat, K.P.L., Kosmeder, J.W., 2nd, and Pez-zuto, J.M. (2001) Biological effects of resver-atrol Antioxid Redox Signal 3, 1041–64.

7. Olas, B., Wachowicz, B., Saluk-Juszczak, J.,and Zielinski, T. (2002) Effect of resveratrol,a natural polyphenolic compound, on plateletactivation induced by endotoxin or thrombinThromb Res 107, 141–5.

8. Wang, Z., Zou, J., Huang, Y., Cao, K., Xu,Y., and Wu, J.M. (2002) Effect of resveratrolon platelet aggregation in vivo and in vitroChin Med J (Engl) 115, 378–80.

9. Kris-Etherton, P.M., Hecker, K.D.,Bonanome, A., Coval, S.M., Binkoski, A.E.,Hilpert, K.F., Griel, A.E., and Etherton, T.D.(2002) Bioactive compounds in foods: theirrole in the prevention of cardiovascular dis-ease and cancer Am J Med 113(Suppl 9B),71–88S.

10. Deana, R., Turetta, L., Donella-Deana, A.,Dona, M., Brunati, A.M., De Michiel,L., and Garbisa, S. (2003) Green tea

238 Mousa

epigallocatechin-3-gallate inhibits plateletsignalling pathways triggered by both pro-teolytic and non-proteolytic agonists ThrombHaemost 89, 866–74.

11. Kang, W.S., Chung, K.H., Chung, J.H., Lee,J.Y., Park, J.B., Zhang, Y.H., Yoo, H.S.,and Yun, Y.P. (2001) Antiplatelet activity ofgreen tea catechins is mediated by inhibitionof cytoplasmic calcium increase J CardiovascPharmacol 38, 875–84.

12. Kang, W.S., Lim, I.H., Yuk, D.Y., Chung,K.H., Park, J.B., Yoo, H.S., and Yun, Y.P.(1999) Antithrombotic activities of greentea catechins and (-)-epigallocatechin gallateThromb Res 96, 229–37.

13. Kazi, A., Smith, D.M., Daniel, K., Zhong,S., Gupta, P., Bosley, M.E., and Dou, Q.P.(2002) Potential molecular targets of teapolyphenols in human tumor cells: signif-icance in cancer prevention In Vivo 16,397–403.

14. Sachinidis, A., Seul, C., Seewald, S., Ahn, H.,Ko, Y., and Vetter, H. (2000) Green tea com-pounds inhibit tyrosine phosphorylation ofPDGF beta-receptor and transformation ofA172 human glioblastoma FEBS Lett 471,51–5.

15. Ali, M., Afzal, M., Gubler, C.J., and Burka,J.F. (1990) A potent thromboxane formationinhibitor in green tea leaves ProstaglandinsLeukot Essent Fatty Acids 40, 281–3.

16. Hodgson, J.M., Puddey, I.B., Burke, V.,Beilin, L.J., Mori, T.A., and Chan, S.Y.(2002) Acute effects of ingestion of blacktea on postprandial platelet aggregation inhuman subjects Br J Nutr 87, 141–5.

17. Kondo, K., Suzuki, Y., Ikeda, Y., andUmemura, K. (2002) Genistein, anisoflavone included in soy, inhibits throm-botic vessel occlusion in the mouse femoralartery and in vitro platelet aggregation Eur JPharmacol 455, 53–7.

18. Ozaki, Y., Yatomi, Y., Jinnai, Y., and Kume,S. (1993) Effects of genistein, a tyrosinekinase inhibitor, on platelet functions. Genis-tein attenuates thrombin-induced Ca2+mobilization in human platelets by affect-ing polyphosphoinositide turnover BiochemPharmacol 46, 395–403.

19. Liu, W., Song, Z.J., and Liang, N.C. (1998)Effects of genistein on aggregation andcytosolic free calcium in pig platelets Zhong-guo Yao Li Xue Bao 19, 540–2.

20. McNicol, A. (1993) The effects of genisteinon platelet function are due to thromboxanereceptor antagonism rather than inhibition oftyrosine kinase Prostaglandins Leukot EssentFatty Acids 48, 379–84.

21. Chinni, S.R., Alhasan, S.A., Multani,A.S., Pathak, S., and Sarkar, F.H. (2003)Pleotropic effects of genistein on MCF-7breast cancer cells Int J Mol Med 12, 29–34.

22. Myoung, H., Hong, S.P., Yun, P.Y., Lee,J.H., and Kim, M.J. (2003) Anti-cancereffect of genistein in oral squamous cell car-cinoma with respect to angiogenesis and invitro invasion Cancer Sci 94, 215–20.

23. Belisario, M.A., Tafuri, S., Di Domenico, C.,Della Morte, R., Squillacioti, C., Lucisano,A., and Staiano, N. (2000) Immobilisedechistatin promotes platelet adhesion andprotein tyrosine phosphorylation BiochimBiophys Acta 1497, 227–36.

24. Kumar, C.C., Nie, H., Rogers, C.P.,Malkowski, M., Maxwell, E., Catino, J.J.,and Armstrong, L. (1997) Biochemical char-acterization of the binding of echistatin tointegrin alphavbeta3 receptor J PharmacolExp Ther 283, 843–53.

25. Staiano, N., Della Morte, R., Di Domenico,C., Tafuri, S., Squillacioti, C., Belisario,M.A., and Di Natale, P. (1997) Echis-tatin inhibits pp72syk and pp125FAK phos-phorylation in fibrinogen-adherent plateletsBiochimie 79, 769–73.

26. Tan, L., Kowalska, M.A., Romo, G.M.,Lopez, J.A., Darzynkiewicz, Z., andNiewiarowski, S. (1999) Identification andcharacterization of endothelial glycoproteinIb using viper venom proteins modulatingcell adhesion Blood 93, 2605–16.

27. Chiu, P.J., Vemulapalli, S., Chintala, M.,Kurowski, S., Tetzloff, G.G., Brown, A.D.,and Sybertz, E.J. (1997) Inhibition ofplatelet adhesion and aggregation by E4021,a type V phosphodiesterase inhibitor, inguinea pigs Naunyn Schmiedeberg’s ArchPharmacol 355, 463–9.

28. Della Morte, R., Squillacioti, C., Garbi, C.,Derkinderen, P., Belisario, M.A., Girault,J.A., Di Natale, P., Nitsch, L., and Staiano,N. (2000) Echistatin inhibits pp125FAKautophosphorylation, paxillin phosphoryla-tion and pp125FAK-paxillin interaction infibronectin-adherent melanoma cells Eur JBiochem 267, 5047–54.

29. Puri, R.N., and Colman, R.W. (1993)Thrombin- and cathepsin G-induced plateletaggregation: effect of protein kinase Cinhibitors Anal Biochem 210, 50–7.

30. Tzeng, S.H., Ko, W.C., Ko, F.N., and Teng,C.M. (1991) Inhibition of platelet aggre-gation by some flavonoids Thromb Res 64,91–100.

31. Fotsis, T., Pepper, M.S., Montesano, R.,Aktas, E., Breit, S., Schweigerer, L., Rasku,


S., Wahala, K., and Adlercreutz, H. (1998)Phytoestrogens and inhibition of angiogen-esis Baillieres Clin Endocrinol Metab 12,649–66.

32. Lee, W.R., Shen, S.C., Lin, H.Y., Hou, W.C.,Yang, L.L., and Chen, Y.C. (2002) Wogo-nin and fisetin induce apoptosis in humanpromyeloleukemic cells, accompanied by adecrease of reactive oxygen species, and acti-vation of caspase 3 and Ca(2+)-dependentendonuclease Biochem Pharmacol 63,225–36.

33. Blania, G., and Spangenberg, B. (1991) For-mation of allicin from dried garlic (Alliumsativum): a simple HPTLC method for simul-taneous determination of allicin and ajoenein dried garlic and garlic preparations PlantaMed 57, 371–5.

34. Makheja, A.N., and Bailey, J.M. (1990)Antiplatelet constituents of garlic and onionAgents Actions 29, 360–3.

35. Mayeux, P.R., Agrawal, K.C., Tou, J.S., King,B.T., Lippton, H.L., Hyman, A.L., Kad-owitz, P.J., and McNamara, D.B. (1988)The pharmacological effects of allicin, a con-stituent of garlic oil Agents Actions 25,182–90.

36. Rendu, F., Daveloose, D., Debouzy, J.C.,Bourdeau, N., Levy-Toledano, S., Jain,M.K., and Apitz-Castro, R. (1989) Ajoene,the antiplatelet compound derived from gar-lic, specifically inhibits platelet release reac-tion by affecting the plasma membrane inter-nal microviscosity Biochem Pharmacol 38,1321–8.

37. Srivastava, K.C., and Tyagi, O.D. (1993)Effects of a garlic-derived principle (ajoene)on aggregation and arachidonic acidmetabolism in human blood plateletsProstaglandins Leukot Essent Fatty Acids 49,587–95.

38. Hirsch, K., Danilenko, M., Giat, J., Miron,T., Rabinkov, A., Wilchek, M., Mirelman,D., Levy, J., and Sharoni, Y. (2000) Effectof purified allicin, the major ingredient offreshly crushed garlic, on cancer cell prolif-eration Nutr Cancer 38, 245–54.

39. Ahmed, N., Laverick, L., Sammons, J.,Zhang, H., Maslin, D.J., and Hassan,H.T. (2001) Ajoene, a garlic-derived naturalcompound, enhances chemotherapy-inducedapoptosis in human myeloid leukaemiaCD34-positive resistant cells Anticancer Res21, 3519–23.

40. Bordia, A., Verma, S.K., and Srivastava, K.C.(1998) Effect of garlic (Allium sativum) onblood lipids, blood sugar, fibrinogen and fib-rinolytic activity in patients with coronary

artery disease Prostaglandins Leukot EssentFatty Acids 58, 257–63.

41. Qi, R., Liao, F., Inoue, K., Yatomi, Y., Sato,K., and Ozaki, Y. (2000) Inhibition by dial-lyl trisulfide, a garlic component, of intracel-lular Ca(2+) mobilization without affectinginositol-1,4, 5-trisphosphate (IP(3)) forma-tion in activated platelets Biochem Pharmacol60, 1475–83.

42. Li, Y., and Lu, Y.Y. (2002) Isolation of dial-lyl trisulfide inducible differentially expressedgenes in human gastric cancer cells by modi-fied cDNA representational difference analy-sis DNA Cell Biol 21, 771–80.

43. Feng, Z.H., Zhang, G.M., Hao, T.L., Zhou,B., Zhang, H., and Jiang, Z.Y. (1994) Effectof diallyl trisulfide on the activation of Tcell and macrophage-mediated cytotoxicity JTongji Med Univ 14, 142–7.

44. Freedman, J.E., and Keaney, J.F., Jr. (2001)Vitamin E inhibition of platelet aggregationis independent of antioxidant activity J Nutr131, 374–7S.

45. Steiner, M. (1999) Vitamin E, a modifier ofplatelet function: rationale and use in car-diovascular and cerebrovascular disease NutrRev 57, 306–9.

46. Mabile, L., Bruckdorfer, K.R., and Rice-Evans, C. (1999) Moderate supplementa-tion with natural alpha-tocopherol decreasesplatelet aggregation and low-density lipopro-tein oxidation Atherosclerosis 147, 177–85.

47. Steiner, M. (1991) Influence of vitamin E onplatelet function in humans J Am Coll Nutr10, 466–73.

48. Zbikowska, H.M., and Olas, B. (2000)Antioxidants with carcinostatic activity(resveratrol, vitamin E and selenium) in mod-ulation of blood platelet adhesion J PhysiolPharmacol 51, 513–20.

49. Suttuar, J., Masova, L., and Scheiner, T.(2002) Role of free radicals in blood plateletactivation Cas Lek Cesk 141(Suppl.)

50. Kakishita, E., Suehiro, A., Oura, Y., andNagai, K. (1990) Inhibitory effect of vitaminE (alpha-tocopherol) on spontaneous plateletaggregation in whole blood Thromb Res 60,489–99.

51. Toivanen, J.L. (1987) Effects of selenium,vitamin E and vitamin C on human prosta-cyclin and thromboxane synthesis in vitroProstaglandins Leukot Med 26, 265–80.

52. Bakaltcheva, I., Gyimah, D., and Reid,T. (2001) Effects of alpha-tocopherol onplatelets and the coagulation system Platelets12, 389–94.

53. Ricciarelli, R., Zingg, J.M., and Azzi, A.(2002) The 80th anniversary of vitamin E:

240 Mousa

beyond its antioxidant properties Biol Chem383, 457–65.

54. McCarty, M.F. (1986) An antithromboticrole for nutritional antioxidants: implicationsfor tumor metastasis and other pathologiesMed Hypotheses 19, 345–57.

55. Corl, C.M., Cao, Y.Z., Cohen, Z.S., and Sor-dillo, L.M. (2003) Oxidant stress enhancesLyso-PAF-AcT activity by modifying phos-pholipase D and phosphatidic acid in aorticendothelial cells Biochem Biophys Res Com-mun 302, 610–4.

56. Hampel, G., Watanabe, K., Weksler, B.B.,and Jaffe, E.A. (1989) Selenium deficiencyinhibits prostacyclin release and enhancesproduction of platelet activating factor byhuman endothelial cells Biochim Biophys Acta1006, 151–8.

57. Perona, G., Schiavon, R., Guidi, G.C.,Veneri, D., and Minuz, P. (1990) Seleniumdependent glutathione peroxidase: a physio-logical regulatory system for platelet functionThromb Haemost 64, 312–8.

58. Ricetti, M.M., Guidi, G.C., Tecchio, C., Bel-lisola, G., Rigo, A., and Perona, G. (1999)Effects of sodium selenite on in vitro interac-tions between platelets and endothelial cellsInt J Clin Lab Res 29, 80–4.

59. Vitoux, D., Chappuis, P., Arnaud, J., Bost,M., Accominotti, M., and Roussel, A.M.(1996) Selenium, glutathione peroxidase,peroxides and platelet functions Ann BiolClin (Paris) 54, 181–7.

60. Brtkova, A. (1992) Biological effects of sele-nium Bratisl Lek Listy 93, 629–32.

61. Costa, A., Sacchini, V., and Decensi, A.(1993) Retinoids and tamoxifen in breastcancer chemoprevention Int J Clin Lab Res23, 53–5.

62. Parcell, S. (2002) Sulfur in human nutritionand applications in medicine Altern Med Rev7, 22–44.

63. Barrager, E., Veltmann, J.R., Jr., Schauss,A.G., and Schiller, R.N. (2002) A multicen-tered, open-label trial on the safety and effi-cacy of methylsulfonylmethane in the treat-ment of seasonal allergic rhinitis J AlternComplement Med 8, 167–73.

64. Ebisuzaki, K. (2003) Aspirin and methyl-sulfonylmethane (MSM): a search for com-mon mechanisms, with implications forcancer prevention Anticancer Res 23,453–8.

65. Thrombosis prevention – disease therapiesprotocol. Life extension (Accessed at http://www.lef.org/protocols/prtcl-155.shtml.)

66. Baruch, D., Franssen, J., Hemker, H.C.,and Lindhout, T. (1985) Effect of hep-arin and low molecular weight heparins onthrombin-induced blood platelet activationin the absence of antithrombin III ThrombRes 38, 447–58.

67. Marsh Lyle, E., Lewis, S.D., Lehman, E.D.,Gardell, S.J., Motzel, S.L., and Lynch, J.J.,Jr. (1998) Assessment of thrombin inhibitorefficacy in a novel rabbit model of simultane-ous arterial and venous thrombosis ThrombHaemost 79, 656–62.

68. Atagi, S., Sone, S., Fukuta, K., and Ogura,T. (1992) Inhibition by fibrin coagulationof lung cancer cell destruction by humaninterleukin-2-activated killer cells Jpn J Can-cer Res 83, 1088–94.

69. Weitz, J.I., and Crowther, M. (2002)Direct thrombin inhibitors Thromb Res 106,V275–84.

70. Warkentin, T.E. (1997) Heparin-inducedthrombocytopenia. Pathogenesis, frequency,avoidance and management Drug Saf 17,325–41.

71. Markwardt, F. (2002) Hirudin as alterna-tive anticoagulant – a historical review SeminThromb Hemost 28, 405–14.

72. Guillin, M.C. (1998) Thrombin and itspharmacologic regulation Presse Med 27(Suppl 2), 22–7.

73. Shen, G.X. (2001) Development and currentapplications of thrombin-specific inhibitorsCurr Drug Targets Cardiovasc Haematol Dis-ord 1, 41–9.

74. Eikelboom, J.W., and French, J. (2002)Management of patients with acute coronarysyndromes: what is the clinical role of directthrombin inhibitors? Drugs 62, 1839–52.

75. Bode, C., Hanson, S.R., Schmedtje, J.F., Jr.,Haber, E., Mehwald, P., Kelly, A.B., Harker,L.A., and Runge, M.S. (1997) Antithrom-botic potency of hirudin is increased in non-human primates by fibrin targeting Circula-tion 95, 800–4.

76. Bossavy, J.P., Sakariassen, K.S., Rubsamen,K., Thalamas, C., Boneu, B., and Cadroy,Y. (1999) Comparison of the antithrom-botic effect of PEG-hirudin and heparin ina human ex vivo model of arterial thrombosisArterioscler Thromb Vasc Biol 19, 1348–53.

Chapter 10

Assessment of Anti-Metastatic Effects of Anticoagulantand Antiplatelet Agents Using Animal Modelsof Experimental Lung Metastasis

Ali Amirkhosravi, Shaker A. Mousa, Mildred Amaya, Todd Meyer,Monica Davila, Theresa Robson, and John L. Francis

Abstract

It is well established that the blood coagulation system is activated in cancer. In addition, there isconsiderable evidence to suggest that clotting activation plays an important role in the biology of malig-nant tumors, including the process of blood-borne metastasis. For many years our laboratory has usedexperimental models of lung metastasis to study the events that follow the introduction of procoagulant-bearing tumor cells into circulating blood. This chapter focuses on the basic methods involved in assessingthe anti-metastatic effects of anticoagulants and anti-platelet agents using rodent models of experimentalmetastasis. In addition, it summarizes our experience with these models, which collectively suggests thatintravascular coagulation and platelet activation are a necessary prelude to lung tumor formation and thatinterruption of coagulation pathways or platelet aggregation may be an effective anti-metastatic strategy.

Key words: Anticoagulants, metastasis, experimental models, tissue factor, anti-platelet drugs.

1. Introduction

1.1. CoagulationActivation in Cancer

Malignancy is associated with activation of the coagulation sys-tem (1) and patients with cancer are at significantly increasedrisk of developing venous thrombosis (2). A link between can-cer and thrombosis was first recognized by Bouillaud in 1823(3). In 1865, Armand Trousseau observed a high incidence ofthromboembolic disease (TED) in patients with gastric carci-noma and further suggested that thrombophlebitis may be symp-tomatic of cancer of other internal organs (4). Substantial clinical,laboratory, pharmacological, and histological evidence has since

S. A. Mousa (ed.), Anticoagulants, Antiplatelets, and Thrombolytics, Methods in Molecular Biology 663,DOI 10.1007/978-1-60761-803-4_10, © Springer Science+Business Media, LLC 2003, 2010

241

242 Amirkhosravi et al.

been accumulated in support of this relationship. Currently, theterm Trousseau’s syndrome is used to describe any type of coagu-lopathy associated with malignancy ranging from venous and/orarterial thromboembolism to chronic disseminated intravascularcoagulation (DIC) with or without thrombotic microangiopathyor secondary hyperfibrinogenolysis (5, 6).

It is now well established that many cancer patients exhibita hypercoagulable state or low-grade DIC. In general, clottingchanges are rarely marked, usually asymptomatic, and more com-monly associated with diffuse thrombosis rather than bleeding.The most common hemostatic abnormalities are hyperfibrinogen-emia and thrombocytosis which are found in 50–80% and 10–57%of cancer patients, respectively (7–10).

Thrombocytopenia is less common and usually occurs due tocytotoxic therapy or bone marrow involvement. Routine coagu-lation screening tests such as prothrombin time (PT) and acti-vated partial thromboplastin time (APTT) are not usually help-ful in demonstrating a hypercoagulable state in cancer patients.Unequivocal detection of hypercoagulability can be achieved withsensitive molecular markers of coagulation activation such as fib-rinopeptide A (FpA) or thrombin–antithrombin (TAT) complex.Elevated levels of these markers are observed in up to 90% ofcancer patients (11). Interestingly, FpA levels increase as thepatients become terminally ill and persistently elevated levels mayindicate treatment failure and a poor prognosis (12). Finally,many patients have an increased rate of coagulation that can bedetected with relatively simple whole blood coagulation tests suchas Thromboelastography R© (TEG) and SonoclotTM analysis (13).

1.2. The Roleof Tissue Factorin Cancer-AssociatedCoagulationActivation

Although the mechanisms by which coagulation is activated dur-ing malignancy are multifactorial, tissue factor (TF), the primaryinitiator of coagulation, has been recognized to play an impor-tant role in this process. Aberrant expression of TF has beenassociated with various pathological conditions, including can-cer. Many tumor cell types constitutively express TF on theirsurfaces and may trigger the production of TF by adjacent hostcells (monocytes and endothelial cells) (14). Physiologically, TFexpression is limited to extravascular sites such as subendothe-lial layers of the vessel wall. However, two forms of circulatingTF have been described: an alternatively spliced soluble protein(15), and a form associated with cell-derived microparticles (MP)(16). The contribution of the latter type of TF to a prothrom-botic state in cancer has been suggested. High levels of circu-lating TF have been found in the plasma of patients with differ-ent cancer types and correlated with other coagulation activationmarkers (17, 18). A recent case study showed extremely high lev-els of MP-associated TF in the plasma of a lung cancer patientwith a severe form of Trousseau’s syndrome (19). Although TF

Assessment of Anti-Metastatic Effects of Anticoagulant and Antiplatelet Agents 243

activity in the patient’s plasma did not correlate with the high lev-els of circulating TF antigen, a recent study from the authors’ lab-oratory showed that TF-dependent procoagulant activity was cor-related with TAT levels in cell-free plasmas of mice with growingorthotopic pancreatic human tumors, supporting the hypothesisthat circulating tumor-derived TF causes coagulation activation invivo (20).

1.3. The Roleof Tissue Factorand Thrombinin Tumor Growth,Angiogenesis,and Metastasis

The effects of cancer on the coagulation protease cascade donot constitute a unidirectional relationship. Rather, a large bodyof evidence suggests that components of the hemostatic systempromote tumor growth, angiogenesis, and metastasis, suggest-ing that activation of the clotting pathways in cancer patientsis not merely an epiphenomenon of the disease. One potentialpathway involves TF-mediated clotting activation, which leads tothrombin generation (locally or systemically), platelet activation,and tumor-associated fibrin formation. TF also appears to reg-ulate, directly or indirectly, tumor angiogenesis via mechanismsindependent of clotting activation. In addition, the extracellu-lar functions of the TF/VIIa complex appear to cooperate withthe signaling functions of the TF cytoplasmic domain to supportblood-borne metastasis (21, 22). The information currently avail-able on the multiple effects of the TF pathway on tumor patho-physiology provides the basis for considering TF as a target foranti-metastatic and anti-angiogenic therapy.

Thrombin is a multi-functional serine protease that is rapidlygenerated following clotting activation. In addition to its directrole in fibrin formation and platelet activation, which enhancetumor metastasis, thrombin can promote angiogenesis via directas well as indirect mechanisms. For example, thrombin canpotentiate vascular endothelial growth factor (VEGF) activity onendothelial cells by up-regulating the expression of VEGF recep-tors on endothelial cells (23). Both thrombin and VEGF can inturn stimulate DNA synthesis in endothelial cells, either alone, orin a synergistic manner (24).

1.4. The Role of theCoagulation Systemin Blood-BorneMetastasis

For many years, research in the authors’ laboratory has focusedon the events that take place when cancer cells enter the circulat-ing blood (11). Specifically, we have investigated whether coagu-lation activation induced directly by tumor cells is important inhematogenous (blood-borne) metastasis. To address this issue,we have utilized experimental models of metastasis. In this typeof model, TF-expressing tumor cells are injected intravenouslyinto the tail vein of mice or rats. Injected tumor cells immedi-ately become entrapped in the microvasculature of the lung – “theorgan of first encounter.” Consequently, evidence of intravascu-lar coagulation can be detected shortly after tumor cell injection.This includes a fall in platelet count and fibrinogen and Factor


X (FX) levels and the presence of increased plasma hemoglobinlevels. The latter heralds the onset of microangiopathic hemolyticanemia caused by formation of fibrin strands in the microvascu-lature. Animals injected with tumor cells subsequently developsecondary tumors in the lungs. It is important to note that ani-mal models of experimental metastasis do not completely repre-sent human metastasis, since they do not include stages of themetastatic processes that occur prior to tumor cell entrance intothe blood stream (e.g., migration, invasion, and intravasation).However, these models serve to isolate the events of interest inthe circulating blood and thus are useful for proof of conceptstudies. In this current chapter, we aim to: (a) describe a numberof methods that are useful in assessing the effects of anticoagu-lants or anti-platelet agents on tumor cell-induced clotting acti-vation and experimental lung metastasis; (b) describe the resultsobtained from various such agents and provide some perspectiveon how such data could be interpreted.

2. Materialsand Methods

2.1. Tumor Cells Human or murine tumor cells are cultured in their appropriatemedia until 70% confluent. Cells are then harvested, preferablywith a non-enzymatic cell dissociation solution (Sigma), washed,and suspended in phosphate-buffered saline (PBS) at the desiredconcentrations. It is important for the final cell suspension to befree of visible aggregates.

2.2. IntravenousInjections

The recommended intravenous injection volume of cell suspen-sions or reagents is 0.1–0.2 ml for mice, although up to 0.5 mlcould be injected slowly (∼ 30 s). Before injections, animals arewarmed for 3–5 min using a heat lamp. They are then placedin a standard restrainer and are injected in the lateral tail vein(Fig. 10.1a). In our studies, the number of tumor cells injectedin one animal ranged from 1×105 to 2×106 cells. Typically, inorder to observe marked tumor cell-induced clotting activation,1–2×106 cells were injected. However, for experimental metasta-sis studies, 1×105 – to 1×106 cells were given.

2.3. Blood Draws Blood samples are obtained by cardiac puncture of anesthetizedanimals (Fig. 10.1b). Exactly 0.5 ml of blood is collected into a1-ml syringe containing 0.1 ml of 3.2% trisodium citrate using a27- or 25-gauge needle for mice or rats, respectively.

2.4. Evidence ofClotting Activation

In our studies we used platelet count, FX and fibrinogen con-centrations, and plasma hemoglobin levels as indicators of tumor


Fig. 10.1. (a) Injection of tumor cells into the tail vein of experimental mice. Cells injected in this manner are arrestedrapidly in the microvasculature of the lungs – the organ of first encounter – and activate blood coagulation. (b) Blood(0.5 ml) is collected by cardiac puncture into a syringe containing trisodium citrate.

cell-induced clotting activation. Platelet counts were measuredautomatically using an electronic Coulter Counter no later than10 min after blood draws. The blood was then centrifuged at5,000 rpm for 10 min. The cell-free plasma was separated andfrozen at –70◦C until analysis. FX and fibrinogen levels weremeasured by standard automated coagulation techniques. Plasmahemoglobin levels were measured by the orthotolidine colori-metric assay (25) modified for use in microtitre plates. A rangeof hemoglobin concentrations between 5 and 100 μg/ml wasprepared by dilution of a standard hemoglobin solution (BDHChemicals). Animal plasmas were diluted 1:5 in PBS. To eachwell of a microtitre plate were added 130 μl of orthotolidinesolution (0.25% [w/v] orthotolidine in 90% [v/v] glacial aceticacid) and 5 μl of diluted standard solution or test plasma. Thereagents were mixed and after 2 min, 130 μl of 1.2% hydro-gen peroxide were added to each well. After 10 min at roomtemperature, 40 μl of glacial acetic acid were finally added andabsorbances were immediately measured at 630 nm using a platereader.

2.5. Quantitation ofPulmonary TumorNodules

Fourteen to forty-two days (depending on the cell line used) aftertumor cell injection, animals were euthanized and the lungs weredissected en bloc from the thoracic cage, rinsed in water, andplaced in Bouin’s fixative solution (Sigma). On occasions whenthere are many surface nodules present, enumeration of tumorfoci can be made easier if the lungs are insufflated with the fix-ative. Briefly prior to dissection of the lungs, 2–3 ml of Bouin’ssolution is injected via the trachea, thus insufflating the lungs. Thelungs are then dissected out as above, rinsed to remove excess fix-ative, and evaluated macroscopically.


3. Summaryof the ResultsObtained fromModels ofExperimentalMetastasis

3.1. Effectsof Heparinsand Coumadin

We first set out to assess the effects of the commonly used anti-coagulants heparin and coumadin in a rat model of experimen-tal metastasis (26–29). Unfractionated heparin was administeredintravenously 1 h before intravenous injection of MC28 fibrosar-coma cells. Coumadin was given in the animals’ drinking water,starting 1 week before tumor cells were injected to allow anti-coagulation to achieve therapeutic levels (PT at least twice thatat baseline). In control (no anticoagulant therapy) animals, evi-dence of tumor cell-induced coagulation was observed within30 min of tumor cell injection. These changes were almost com-pletely suppressed in anticoagulated animals. In addition, systemicanticoagulation significantly reduced lung tumor formation com-pared with controls. Delaying warfarinization almost abolishedthe antimetastatic effect of coumadin, suggesting that the impor-tant effects in this model occur at an early stage post-tumor injec-tion (30).

We then investigated whether the anti-metastatic effect ofanticoagulants was due to inhibition of tumor cell growth in thelungs or whether anticoagulation reduced the physical trapping oftumor cells in the pulmonary microvasculature. To address this,we conducted fate studies of radiolabeled tumor cells in controland anticoagulated animals (29). The results showed that the anti-metastatic effects of heparin and coumadin in this model werenot due to inhibition of initial tumor cell trapping in the lung,but were due to increased subsequent clearance of the cells fromthe pulmonary microvasculature. These data indicated that, in thepresence of competent coagulation pathways, tumor cells triggerthrombin generation and form complexes with platelets and fib-rin that potentially aid adherence to the vascular endothelium, anevent that is a pre-requisite to the important metastatic step ofextravasation. Anticoagulation reduced the ability of tumor cellsto activate coagulation and platelets and hence to form stickyaggregates. Thus they were cleared more rapidly from the lung,before they had the opportunity to form tumors. These eventsare summarized in Table 10.1. Our results are consistent withfindings of Palumbo and colleagues who demonstrated a reducedretention of radiolabeled tumor cells injected into fibrinogen-deficient mice (31). Interestingly, these authors further demon-strated that inhibition of thrombin by hirudin was found to


Table 10. 1Effect of anticoagulation on the intravascular eventsfollowing tail vein injection of tumor cells in experimentalanimals

Normal coagulation Anticoagulated

Cell trapping Cell trapping

Platelet aggregation No platelet aggregationFibrin formation No fibrin formation

Delayed clearance Rapid clearanceExtravasation Cell death

Tumor formation No tumor formation

further reduce the already low metastatic potential of tumorcells in fibrinogen-deficient mice. It was thus concluded that fib-rin(ogen) is an important determinant of metastatic potential andthat thrombin contributes to the metastatic process through atleast one fibrinogen-independent mechanism.

In our study fibrin-platelet microthrombi were observed byelectron and light microscopy 30–60 min after intravenous injec-tion of tumor cells. Fibrin was observed adjacent to tumor cells,but it was not clear from these experiments whether fibrin gener-ation occurred on the surface of fibrosarcoma cells. Rapid accu-mulation of radiolabeled platelets at the site of tumor lodgementsupported the microscopic observations.

3.2. Anti-metastaticEffect of aNon-anticoagulantLow-MolecularHeparin (LMWH)Versus the StandardLMWH Enoxaparin

Having observed the anti-metastatic effect of unfractionatedheparin, we then tested the effects of the LMWH enoxaparinas well as a non-anticoagulant low molecular weight heparin(NA-LMWH) on tumor cell-induced clotting activation in vivoand experimental metastasis (32). NA-LMWH was prepared byfragmenting porcine mucosal heparin into LMWH followed byreduction with sodium borohydride and acid hydrolysis. NA-LMWH does not exhibit any inhibitory activity against factorsXa and IIa. However, it has the ability to cause the release oftissue factor pathway inhibitor (TFPI) from vascular endothe-lial cells in vitro similar to enoxaparin (Fig. 10.2). We there-fore hypothesized that NA-LMWH will exhibit only a mod-erate anti-metastatic effect (compared to enoxaparin) withoutaffecting tumor cell-induced clotting activation in vivo. In thisstudy we used the B16 melanoma mouse model of experimentalmetastasis.

The anticoagulant effect of enoxaparin and NA-LMWH(both at 10 mg/kg) was measured 3 h after subcutaneous or15 min after intravenous (tail vein) injection of either drug


Concentration (μg/106 cells)0.01 0.05 0.10 0.50 1.00 3.00 10.00

TF

PI (

ng/1

06 ce

lls)

0

100

200

300

400

NA-LMWH Enoxaparin

Fig. 10.2. Both enoxaparin and a non-anticoagulant low molecular weight heparin(NA-LMWH) have the ability to cause the release of TFPI from endothelial cells in vitro.

in experimental mice. Blood samples were collected by car-diac puncture into trisodium citrate and whole blood coagula-tion was assessed by TEG and by a Sonoclot Analyzer (Sienco,USA). (APTT); anti-Xa, and anti-IIa activity of the heparins inmouse platelet-poor plasmas were also measured using an auto-mated coagulation analyzer. For metastasis experiments, micewere injected subcutaneously with a bolus of saline or enoxa-parin or NA-LMWH and 4 h thereafter all mice were injectedintravenously with B16 tumor cells. The mice then received dailydoses (for 14 days) of either heparin. Experimental lung metasta-sis was assessed 15 days after tumor cell injection. As expected,enoxaparin, but not NA-LMWH, exhibited significant anti-Xaand anti-IIa activities. Intravenous injection of melanoma cellsresulted in a significant and rapid reduction (>50%) in the plateletcount of control mice previously injected with saline. This fall inplatelet count was abolished in mice treated with enoxaparin priorto tumor cell injection. In contrast NA-LMWH had no effect ontumor cell-induced thrombocytopenia (Fig. 10.3). Both enoxa-parin and NA-LMWH reduced experimental metastasis by 70%(P<0.01, Fig. 10.4).

In the context of other interventional studies involving theanticoagulants mentioned above, the significant anti-metastaticeffect of NA-LMWH is intriguing and points to anti-metastaticmechanisms that are independent of coagulation and platelet acti-vation pathways. The anionic properties of heparin and its frac-tioned derivatives are thought to be responsible for heparins’anti-cancer effects, including angiogenesis, tumor cell adhesion,


0

200

400

600

800

1000

1200

Baseline Control Enoxaparin NA-LMWH

Pla

tele

t n

um

ber

s (x

106 /

ml)

P < 0.01P < 0.01

Fig. 10.3. The effect of enoxaparin versus NA-LMWH on tumor cell-induced thrombo-cytopenia. Unlike enoxaparin, due to its lack of anticoagulant activity, NA-LMWH did notreverse the fall in platelet count caused by the injection of tumor cells.

PBS Enoxaparin NA-LMWH

# o

f lu

ng

tu

mo

r n

od

ule

s

0

50

100

150

200

PBS

ENOXAPARIN

NA-LMWH

P < 0.01P < 0.01

Fig. 10.4. Right panel: The effect of enoxaparin versus NA-LMWH on experimental lung metastasis. Left panel: Repre-sentative lungs of control and heparin-treated mice 2 weeks after the injection of B16 tumor cells.

and malignant transformation. Possible coagulation-independentmechanisms for inhibition include binding of heparin to angio-genic growth factors (such as basic fibroblast growth factor andVEGF) and modulation of TF (33–35). A key component ofmetastasis is the adhesion of circulating tumor cells to the vas-cular endothelium of organs distant from the primary tumor site.P-selectin-mediated tumor cell interactions have been shown topromote metastasis (36, 37). Expression of carbohydrate moi-eties required for P-selectin binding is associated with increasedmetastasis and poor survival in various tumor cell types (38–40).


Furthermore, it has been demonstrated that heparins (unfraction-ated and low molecular weight) and derivatives (fondaparinux)exert differential anti-metastatic effects at comparable anticoagu-lant activities in vivo as well as differential inhibitory effects onP-selectin-mediated interactions (37). It may therefore be possi-ble to suggest that the degree of inhibition of P-selectin functionof a given anticoagulant may provide a parameter to indicate itsanti-metastatic potential.

3.3. The Effect of TFInhibition onExperimentalMetastasis

In order to examine the role of tumor cell TF on experimen-tal metastasis, we used the monoclonal antibody H36 (SunolMolecular Inc, Miramar, FL), which competitively inhibits humanTF/VIIa-dependent FX activation (11). In this study we injectedintravenously the TF-expressing C15 (metastatic) variant of A375human melanoma cells in athymic nude mice (nu/nu). The anti-body was diluted to concentrations of 0.1–3 mg/ml, and 0.2 mlof each concentration were injected intravenously 10 min beforeinjection of tumor cells and at several time points (3, 7, 10, 14,17, 21, 28, and 35 days) thereafter. The experiment was termi-nated 42 days after tumor cell injection.

All doses of H36 antibody were well tolerated by the exper-imental animals. No bleeding or any other adverse effects wereobserved as a result of the antibody injection. H36 inhibitedexperimental lung metastasis at all doses tested (Fig. 10.5). Therewas a relationship between the antibody dose and lung tumor for-mation insofar as all eight animals given 0.1 mg/kg (the lowestdose) had metastasis and this group had the highest total number

H36 (mg/kg)

H36Control

Control 0.1 1 3 30

Nu

mb

er o

f L

un

g T

um

ors

10

0

20

40

80

60

100

120

140

Fig. 10.5. Effects of the anti-human TF monoclonal antibody H36 on experimentallung metastasis. Representative lungs taken from control and antibody-treated miceare included. Tumors appear as pale nodules.


of lung nodules. However, only five of nine animals receiving thehighest dose (30 mg/kg) had observable surface metastatic fociand, as a group had the lowest number of nodules. In a previ-ous study, we preincubated the same tumor cells with or withouttwo anti-TF antibodies, H36 and I43 (which blocks the bindingof VII(a) to TF), before injecting the cells into experimental ani-mals. In these experiments, we also demonstrated that tumor cellTF activity blockade is associated with a significant reduction inexperimental metastasis. It is important to note that in our stud-ies, only TF expressed by human tumor cells was inhibited bythe anti-TF antibody, since H36 does not inhibit the function ofmurine TF (data not shown). This explains the absence of anybleeding side effects in our studies despite high doses of anti-body administered. In a human trial, the H36 antibody causedspontaneous minor bleeding in a dose-related manner althoughthe majority of those bleeding episodes were clinically consistentwith platelet-mediated bleeding (41). Recently, Snyder and col-leagues have developed an elegant mouse model in which expres-sion of human TF is under the control of the murine TF promoter(42). These knock-in mice, referred to as TFKI, have human TFexpression similar to murine TF in wild type animals and havenormal hemostasis. Interestingly, when TFKI mice were treatedwith neutralizing anti-human TF antibody, cardiac bleeding wasobserved. However, the TFKI mouse may prove to be usefulin assessing anti-tumor effects of anti-TF antibodies providedthat tumors expressing human TF are introduced in to the hostanimal.

Next we examined the anti-metastatic effect of the natu-ral inhibitor of TF, TFPI, a plasma protein that regulates TF-dependent reactions by neutralizing the catalytic activity of FXaand/or forming a quaternary inhibitory complex with TF, FVIIa,and FXa on the cell membrane (43, 44). TFPI’s anticoagulantfunction appears to be as efficient as unfractionated heparin (45).In our laboratory, we intravenously injected murine recombinantTFPI shortly before injection of B16 tumor cells in mice (46).This almost completely inhibited tumor cell-induced thrombo-cytopenia (Fig. 10.6a) and significantly reduced experimentalmetastasis (Fig. 10.6b).

We then tested the effect of cellular TFPI expression on themetastatic potential of tumor cells. In order to achieve this, wetransfected B16 cells (which do not express TFPI) with murineTFPI. This significantly reduced tumor cells TF-mediated pro-coagulant activity without altering TF antigen expression. Wheninjected intravenously into mice, the transfected cells failed toproduce a significant fall in platelet counts compared to control(non-transfected) or antisense transfected B16 cells. The TFPI-transfected cells also formed significantly fewer lung tumor nod-ules compared to the above-mentioned control cells (Fig. 10.7).


0

50

100

150

200

250

Control rmTFPI

# o

f lu

ng

tu

mo

r n

od

ule

sA

0

200

400

600

800

1000

1200

Pla

tele

t co

un

t (x

106

/ml)

P < 0.01

B

P < 0.01

Baseline Control (post) rmTFPI(post)

Fig. 10.6. (a) Intravenous injection of recombinant murine TFPI (rmTFPI) significantly inhibited tumor cell-induced throm-bocytopenia caused post-injection of B16 melanoma cells. (b) Experimental lung metastasis was significantly inhibitedby rmTFPI.

These results suggested that both circulating and tumor cell-associated TFPI appear to play a role in blood-borne metasta-sis. Consistent with the above findings, Hembrough et al showeda significant reduction in B16 experimental lung metastasis by


0 10 20 30 40 50

Control

AS1AS2

S1

S2

S3

S4

S5

S6

S7

S8

S9

S10

S11

S12

S13

S14

Cel

l lin

e d

esig

nat

ion

Relative TF activity (units)

TFPI +

TFPI −

Fig. 10.7. The effect of TFPI (sense, antisense, vector) transfection of B16 melanoma cells on cellular procoagulantactivity as measured by a clotting assay. Fourteen sense (S), two antisense (AS) and one vector only cell lines wereprepared and compared to untransfected wild-type B16 cells. S10 and AS1 were used for the metastasis experiments.Sense cell lines exhibited much lower procoagulant activity and caused the formation of significantly fewer lung tumornodules (inset).

intraperitoneal injection of human TFPI (47). It is noteworthythat in the latter study TFPI was given daily after tumor cell injec-tion. The authors also showed that another inhibitor of TF/VIIa,the nematode anticoagulant protein rNAPc2, had a similarantimetastatic effect compared with TFPI, whereas rNAP5,another nematode anticoagulant protein that specifically inhibitsFXa did not exhibit significant antimetastatic effects. Theseresults suggest that the proteolytic activity of TF/VIIa may pro-mote experimental metastasis by mechanisms independent of FXaactivation.

3.4. Effect ofAnti-platelet Agentson ExperimentalMetastasis

Evidence suggests that platelets are an integral part of themicrothrombus that enhances the arrest of tumor cells in circula-tion. Platelets interact with certain tumor cell emboli that couldprolong cell survival in the circulation (48). Platelet–tumor cellemboli can also induce downstream ischemic endothelial damage,which could potentially expose adhesive subendothelial matrixfor tumor cell binding and arrest (49). In addition, sequestra-tion of tumor cells by platelets can protect tumor cells fromimmunological host surveillance (50). As described above, tumor


cell-induced platelet activation and aggregation predominantlyrequire prior activation of the coagulation system and throm-bin generation, although other mechanisms independent ofthrombin have also been reported (51). The anti-metastaticeffect of experimental thrombocytopenia (52) and anti-plateletdrugs such as aspirin (53) and dipyridamole (54) providestrong support for a requirement of platelets in hematogenousmetastasis.

Glycoprotein (Gp) IIb/IIIa plays a central role in homotypicplatelet aggregation and is also involved in the heterotypic adhe-sion of tumor cells to platelets. Inhibition of metastasis by theanti-GpIIb/IIIa monoclonal antibody 10E5, which inhibits bind-ing of fibronectin to von Willebrand factor to platelet GpIIb/IIIa,was first reported by Karpatkin and colleagues (55). In our labora-tory, using a Lewis lung carcinoma (LL2) model of experimentalmetastasis in rats, we examined the effect of GpIIb/IIIa blockadeon lung seeding. In this study, we utilized a murine F(ab’)2 ver-sion of abciximab (7E3), since the human antibody cross reactswith rat, but not mouse, GpIIb/IIIa and αvß3. This treatmentinhibited tumor cell-induced thrombocytopenia and significantlyreduced experimental metastasis (56).

In a second study, using the LL2 lung carcinoma cells, thistime in mice, we examined the effect of a potent non-peptide oralGpIIb/IIIa antagonist, XV454, on tumor cell-induced thrombo-cytopenia and experimental metastasis (57). XV454 has a longreceptor-bound lifetime, similar to 7E3, and binds with highaffinity to either activated or non-activated human, baboon,mouse or canine platelets (Kd = 0.5 nM) (58). Maximal aggrega-tion inhibition by XV454 occurs at approximately 75% receptoroccupancy. XV454 (5 mg/kg) was administered orally or intra-venously 3 h or 10 min prior to tumor cell injection, respec-tively. By whole blood platelet aggregometry, we verified thatmouse platelet aggregation was completely inhibited 1 h afteroral and 10 min after intravenous administration of XV454. Thisinhibitory effect persisted for at least 24 h after oral delivery.Tumor cell-induced thrombocytopenia was inhibited significantlyby both oral and intravenous administration of XV454. Similarly,both oral and intravenous XV454-treated mice had >80% fewersurface lung tumor nodules than the control group (P <0.001,n = 8 per group). However, tumor burden was reduced by 83%in animals receiving intravenous XV454 (P = 0.06), by 50% inthe oral (single pre-tumor cell treatment) group (P = 0.14), andby 91% in the multiple treatment oral group (P = 0.015). Takentogether, the results strongly suggest that platelet activation andaggregation are important for the success of metastases in theseexperimental models.


4. Summary

Our studies with an experimental model of tumor cell-inducedcoagulopathy and lung metastasis collectively suggest that TF isan important determinant of the clotting changes and the abil-ity of tumor cells to form secondary tumors. Our data, and thoseof others, suggest that when procoagulant (TF-expressing) tumorcells enter the circulation, blood coagulation is rapidly activated,particularly at sites of tumor cell arrest. The rapid generation ofthrombin leads to platelet activation and aggregation and theseevents both enhance the metastatic process. Because the primarytrigger for these events is tumor cell TF, the ability to inhibit TF-induced coagulation, either downstream by reducing blood coag-ulability (heparins, coumadin) or more directly through anti-TFantibodies or small molecule TF inhibitors or via TFPI, may bea useful tool in the adjunct treatment of human malignancies.LMWHs may be particularly effective due to their dual effectsof FXa inhibition and elevation of TFPI levels. However, despitepreclinical evidence in favor of LMWH anti-tumor effects, clini-cal trials have reported conflicting results. A recent meta-analysisand systemic review of the efficacy and safety of anticoagulants(unfractionated heparin, LMWH, and warfarin) as cancer treat-ment concluded that these agents, particularly LMWH, signifi-cantly improved the overall survival of cancer patients withoutvenous thrombosis (59). However, the authors also conclude thatat the present time the use of anticoagulants as antineoplastictherapy cannot be recommended until more confirmatory ran-domized controlled trials are performed. Indeed, the translationof the above-described research into effective human therapy mayrequire the development of novel anticoagulants with adequatebioavailability and safety profiles for their long-term use.

5. Notes

1. As mentioned above, the main disadvantage of animal mod-els of experimental metastasis is the fact that large num-bers of tumor cells are introduced directly into the bloodstream. This is non-physiological and the initial steps of themetastatic process leading to intravasation are not consid-ered. In addition, metastasis occurs as a result of the initialtrapping of the tumor cells in the vasculature of the organ of


first encounter (the lungs) and occurs very rapidly. Tumorcell-induced coagulation and platelet activation also occurmore rapidly than in spontaneously metastasizing modelsand may be exaggerated by tumor cell trafficking in themicrovessels. In our experience with several anticoagulantand anti-platelet intervention studies, using this model, anti-hemostatic agents only exert maximal anti-metastatic effectif administered before the injection of tumor cells. Indeed,we have found that delaying the administration of anti-coagulant or anti-platelet agents significantly reduces theiranti-metastatic effect. We believe that this is mainly due tothe fact that the initial arrest and coagulation activation bytumor cells occur minutes after tumor cell injection and thatanticoagulation after this time point does not significantlyimpact the metastatic process.

2. There are two alternatives to the intravenous models ofexperimental metastasis. The first model is known as a“spontaneous” model of metastasis. In these models tumorcells are injected subcutaneously into experimental ani-mals and metastasis to the lung, liver, or other organs isassessed at various time points afterward. Although thismodel represents more closely the metastatic process, for-mation of distant tumor foci occurs very inefficiently. Oftentimes the primary tumor grows to an inappropriately largevolume without any visible signs of metastatic spread. Theremoval of the primary tumor has been shown to enhancemetastatic spread. The second alternative is the use of ortho-topic mouse models. In this model, tumor cells (or tumorfragments) are introduced into their organ of origin. Themain advantage of this type of model is that theoreticallyat least, the growing tumor will follow its natural courseof progression. In this regard, orthotopic models of tumorgrowth and dissemination represent human disease moreclosely than any other model. However, large variations intumor growth patterns present difficulties in metastasis out-comes and data interpretation.

3. There are large variations in the number of pulmonary lungnodules in models of experimental metastasis. In our expe-rience, in control groups we observe pulmonary nodules in100% of injected control animals. However, there are largevariations (e.g., range from 10 to 150) in the number ofsurface tumor foci. For this reason a minimum of 8–10 ani-mals per experimental group is needed to obtain sufficientstatistical rigor.

4. In some experimental metastasis models like the LL2 model,there are large variations in the sizes of the pulmonary tumornodules. In addition, it is possible for multiple surface foci


to coalesce and form apparently much larger nodules. Whencounting the surface nodules, it is recommended to considersmall and large nodules similarly regardless of gross size dif-ferences. However, on these occasions, the measurement oftumor burden by weighing the entire lung tissue provides anadditional and sometimes a more representative parameterfor the evaluation of the anti-metastatic effects of potentialtherapeutic agents.

5. Experimental mice tolerate high daily doses of heparinsand anti-GpIIb/IIIa agents (as highest used in our studies,10–15 mg/kg) without any signs of bleeding. It is impor-tant to note that this tolerance can not be extrapolated tohuman studies. In humans, the therapeutic window of anti-coagulants is certainly narrower than in rodents. This maybe one reason why preclinical animal data on the anti-tumoreffects of anticoagulants are more encouraging than thoseseen in human trials so far.

References

1. De Cicco, M. (2004) The prothromboticstate in cancer: pathogenic mechanisms CritRev Oncol Hematol 50, 187–96.

2. Hillen, H.F. (2000) Thrombosis in cancerpatients Ann Oncol (Suppl. 3) 11, 273–76.

3. Buller, H.R., van Doormaal, F.F., van Sluis,G.L., and Kamphuisen, P.W. (2007) Cancerand thrombosis: from molecular mechanismsto clinical presentations J Thromb Haemost5(Suppl.1), 246–54.

4. Trousseau, A. Phlegmasia alba dolens. In:Baillier, J., ed. Clinique de l’Hotel-Dieu deParis, 2nd edition. London: New SydenhamSociety;1865:654–6.

5. Varki, A. (2007) Trousseau’s syndrome: mul-tiple definitions and multiple mechanismsBlood 110, 723–9.

6. Langer, F., Spath, B., Haubold, K., et al.(2008) Tissue factor procoagulant activity ofplasma microparticles in patients with cancer-associated disseminated intravascular coagu-lation Ann Haematol 87, 451–7.

7. Sun, N., McAfee, W., Hum, G., and WeinerJ.M. (1979) Hemostatic abnormalities inmalignancy, a prospective study of one hun-dred eight patients. Part 1. Coagulation stud-ies Am J Clin Pathol 71, 10–6.

8. Kies, M.S., Posch, J.J., Giolama, J.P., andRubin, R.N. (1980) Hemostatic function incancer patients Cancer 46, 831–7.

9. Edwards, R.L., Rickles, F.R., Moritz, T.E.et al. (1987) Abnormalities of blood coagu-lation tests in patients with cancer Am J ClinPathol 88, 596–602.

10. Davis, R.B., Theologides, A., and Kennedy,B.J. (1969) Comparative studies of bloodcoagulation and platelet aggregation inpatients with cancer and non-malignant dis-eases Ann Int Med 71, 67–80.

11. Francis, J.L., and Amirkhosravi, A. (2002)Effect of antihemostatic agents on experi-mental tumor dissemination Semin ThrombHemost 28, 29–38.

12. Rickles, F.R., Edwards, R.L., Barb, C., andCronlund, M. (1983) Abnormalities of bloodcoagulation in patients with cancer: fib-rinopeptide A generation and tumor growthCancer 51, 301–7.

13. Francis, J.L., Francis, D.A., and Gunathila-gan, G.J. (1994) Assessment of hypercoag-ulability in patients with cancer using thesonoclot analyzer and thromboelastographyThromb Res 74, 335–46.

14. Wojtukiewicz, M.Z., Sierko, E., Klement, P.,and Rak, J. (2001) The hemostatic systemand angiogenesis in malignancy Neoplasia 3,371–84.

15. Bogdanov, V.Y., Balasubramanian, V., Hath-cock, J., et al. (2003) Alternatively splicedhuman tissue factor: a circulating, soluble,thrombogenic protein Nat Med 9, 458–62.

16. Giesen, P.L., Rauch, U., Bohrmann, B., et al.(1999) Blood borne tissue factor: anotherview of thrombosis Proc Natl Acad Sci USA96, 2311–5.

17. Hron, G., Kollars, M., Weber, H., et al.(2007) Tissue factor positive microparti-cles: cellular origin and association with


coagulation activation in patients with col-orectal cancer Thromb Haemost 97, 119–23.

18. Langer, F., Chun, F.K., Amirkhosravi, A.,et al. (2007) Plasma tissue factor antigen inlocalized prostate cancer: distribution, clini-cal significance and correlation with haemo-static activation markers Thromb Haemost 97,464–70.

19. Del conde, I., Bharwani, L.D., Dietzen, D.J.,et al. (2006) Microvesicle-associated tissuefactor and Trousseau’s syndrome J ThrombHaemost 5, 70–74.

20. Davila, M., Amirkhosravi, A., Coll, E., Desai,H., Robles, L., Colon, J., Baker, C.H., andFrancis, J.L. (2008) Tissue factor-bearingmicroparticles derived from tumor cells:impact on coagulation activation ThrombHaemost 6, 1517–24.

21. Bromberg, M.E., Konigsberg, W., Madison,J., et al. (1995) Tissue factor promotesmelanoma metastasis by a pathway indepen-dent of blood coagulation Proc Natl Acad SciUSA 92, 8205–9.

22. Mueller, B.M., and Ruf, W. (1998) Require-ment for binding of catalytically active factorVIIa in tissue factor-dependent experimentalmetastasis J Clin Invest 101, 1372–8.

23. Tsopanoglou, N.E., and Maragoudakis, M.E.(1999) On the mechanism of thrombin-induced angiogenesis. Potentiation of vas-cular endothelial growth factor activity onendothelial cells by upregulation of its recep-tors J Biol Chem 274, 23969–76.

24. Tsopanoglou, N.E., and Maragoudakis. M.E.(2004) Role of thrombin in angiogenesis andtumor progression Semin Thromb Hemost 30,63–9.

25. Lewis, G.P. (1965) Method using ortho-tolidine for the quantitative determination ofhemoglobin in serum and urine J Clin Pathol18, 235–9.

26. Amirkhosravi, M., and Francis, J.L. (1993)Procoagulant effects of the MC28 fibrosar-coma cell line in vitro and in vivo Br JHaematol 85, 736–44.

27. Edwards, R.L., Silver, J., and Rickles, F.R.(1993) Human tumor procoagulants: Reg-istry of the Subcommittee of Haemostasisof the Scientific and Standardization Com-mittee, International Society on Thrombo-sis and Haemostasis Thromb Haemost 69,205–13.

28. Adamson, A.S., Luckert, P., Pollard, M.,et al. (1994) Procoagulant activity may bea marker of the malignant phenotype inexperimental prostate cancer Br J Cancer 69,286–90.

29. Amirkhosravi, M., and Francis, J.L. (1995)Coagulation activation by MC28 fibrosar-

coma cells facilitates lung tumor formationThromb Haemost 73, 59–65.

30. McCulloch, P., and George, W.D. (1987)Warfarin inhibition of metastasis: the role ofanticoagulation Br J Surg 74, 879–83.

31. Palumbo, J.S., and Degen, J.L. (2001) Fib-rinogen and tumor cell metastasis Hemostasis31(Suppl 1), 11–15.

32. Mousa, S.A., Linhardt, R., Francis, J.L.,and Amirkhosravi, A. (2006) Anti-metastaticeffect of a non-anticoagulant low-molecularweight heparin versus the standard low-molecular-weight heparin, enoxaparinThromb Haemost 96, 816–21.

33. Mousa, S.A., and Mohamed, S. (2004) Anti-angiogenic mechanisms and efficacy of thelow molecular weight heparin, tinzaparin:anti-cancer efficacy Oncol Rep 12, 683–8.

34. Mousa, S.A., and Mohamed, S. (2004) Inhi-bition of endothelial cell tube formation bythe low molecular weight heparin tinzaparin,is mediated by tissue factor pathway inhibitorThromb Haemost 92, 627–33.

35. Lupu, C., Poulson, E., Toque feuil, S., et al.(1999) Cellular effects of heparin on theproduction and release of tissue factor path-way inhibitor in human endothelial cells inculture Arterioscler Thromb Vasc Biol 19,2251–62.

36. Kim, Y.J., Borsig, L., Varki, N.M., et al.(1998) P-selectin deficiency attenuatestumor growth and metastasis Proc Natl AcadSci USA 95, 9325–30.

37. Ludwig, R.J., Boehme, B., Podda, M., et al.(2004) Endothelial P-selectin as a target ofheparin action in experimental melanomalung metastasis Cancer Res 64, 2743–50.

38. Nakamori, S., Kameyama, M., Imaoka, S.,et al. (1993) Increased expression of SialylLewis X antigen correlates with poor survivalin patients with colorectal carcinoma: clinico-pathological and immunohistochemical studyCancer Res 53, 32–36.

39. Nakayama, T., Watanabe, M., Katsumata, T.,et al. (1995) Expression of sialyl Lewis(a)as a new prognostic factor for patients withadvanced colorectal carcinoma Cancer 75,2051–6.

40. Jorgensen, T., Berner, A., Kaalhus, O., et al.(1995) Upregulation of the oligosaccharidesialyl Lewis X: a new prognostic parameterin metastatic prostate cancer Cancer Res 55,1817–9.

41. Morrow, D.A., Murphy, S.A., McCabe,C.H., et al. (2005) Potent inhibitionof thrombin with a monoclonal antibodyagainst tissue factor (Sunol-cH36): results ofthe PROXIMATE-TIMI 27 trial Eur Heart26, 682–8.


42. Snyder, L.A., Rudnick, K.A., Tawadros, R.,et al. (2007) Expression of human tissue fac-tor under the control of the mouse tissue fac-tor promoter mediates normal hemostasis inknock-in mice J Thromb Haemost 6, 306–14.

43. Gemmell, C.H., Broze, G.J., Turitto, V.T.,and Nemerson, Y. (1990) Utilization ofa continuous flow reactor to study thelipoprotein–associated coagulation inhibitor(LACI) that inhibits tissue factor Blood 76,2266–71.

44. Bajaj, M.S., Birktoft, J.J., Steer, S.A., andBajaj, S.P. (2001) Structure and biologyof tissue factor pathway inhibitor ThrombHaemost, 86, 959–72.

45. Rapp, J.H., Pan, X.M., Ghermay, A., et al.(1997) A blinded trial of local recombinanttissue factor pathway inhibitor versus eitherlocal or systemic heparin in a vein bypassmodel J Vasc Surg 25, 726–9.

46. Amirkhosravi, A., Meyer, T., Chang, J.Y.,et al. (2002) Tissue factor pathway inhibitorreduces experimental lung metastasis of B16melanoma Thromb Haemost 87, 930–6.

47. Fidler, I.J. (1970) Metastasis: quantitativeanalysis of distribution and fate of tumoremboli labeled with 125 I-5 deoxyuridine JNatl Cancer Inst 45, 773–82.

48. Hembrough, T.A., Swartz, G.M., Pap-athanassiu, A., et al. (2003) Tissue fac-tor/factor VIIa inhibitors block angiogenesisand tumor growth through a nonhemostaticmechanism Cancer Res 63, 2997–3000.

49. Warren, B.A. The micro-injury hypothe-sis and metastasis. In: Honn, K.V., andSloane, B.F., eds. Hemostatic Mechanismsand Metastasis.Boston: M. Nijhoff;1984:56.

50. Nieswandt. B., Hafner, M., Echtenacher, B.,et al. (1999) Lysis of tumor cells by natu-ral killer cells in mice is impeded by plateletsCancer Res 59, 1295.

51. Lerner, W.A., Pearlstein, E., Ambrogio, C.,et al. (1983) A new mechanism for tumor-

induced platelet aggregation: comparisonswith mechanisms shared by other tumorswith possible pharmacologic strategy towardsprevention of metastases Int J Cancer 31,463–9.

52. Gasic, G.J., Gasic, T.B., and Stewart, C.C.(1968) Antimetastatic effects associated withplatelet reduction Proc Natl Acad Sci USA61, 46–52.

53. Gasic, G.J., Gasic, T.B., and Murphy, S.(1972) Anti-metastatic effect of aspirinLancet 2, 934.

54. Tzanakakis, G.N., Agarwal, K.C., and Vez-eridis, M.P. (1993) Prevention of humanpancreatic cancer cell-induced hepatic metas-tasis in nude mice by dipyridamole and itsanalog RA-233 Cancer 71, 2466–71.

55. Karpatkin, S., Pearlstein, E., Ambrogio, C.,et al. (1988) Role of adhesive proteins inplatelet tumor interaction in vitro and metas-tasis formation in vivo J Clin Invest 81,1012–9.

56. Amirkhosravi, A., Amaya, M., Siddiqui, F.,et al. (1999) Blockade of GpIIb/IIIa inhibitsthe release of vascular endothelial growthfactor (VEGF) from tumor cell-activatedplatelets and experimental metastasis Platelets10, 285–92.

57. Amirkhosravi, A., Mousa, S.A., Amaya, M.,et al. (2003) Inhibition of tumor cell-inducedplatelet aggregation and lung metastasis byoral GpIIb/IIIa antagonist XV454 ThrombHaemost 90, 549–54.

58. Mousa, S.A., Forsythe, M., Bozarth, J., et al.(1998) XV454, a novel non-peptide smallmolecule platelet GpIIb/IIIa antagonist withcomparable platelet αIIbß3-binding kineticsto c7E3 J Cardiovasc Pharmacol 32, 736–44.

59. Kuderer, N.M., Khorana, A.A., Lyman,G.H., and Francis, C.W. (2007) A meta-analysis and systematic review of the efficacyand safety of anticoagulants as cancer treat-ment Cancer 110, 1149–61.

Chapter 11

Adhesion Molecules: Potential Therapeutic and DiagnosticImplications

Shaker A. Mousa

Abstract

The role of cell adhesion molecules (CAMs) and extracellular matrix (ECM) proteins in variouspathological processes, including angiogenesis, thrombosis, inflammation, apoptosis, cell migration, andproliferation is well documented. These processes can lead to both acute and chronic disease states suchas ocular diseases, metastasis, unstable angina, myocardial infarction, stroke, osteoporosis, a wide range ofinflammatory diseases, vascular remodeling, and neurodegenerative disorders. A key success in this fieldwas identification of the role of platelet glycoprotein (GP)IIb/IIIa in the prevention and diagnosis of var-ious thromboembolic disorders. The use of soluble adhesion molecules as potential diagnostic markersfor acute and chronic leukocyte, platelet, and endothelial cell insult is becoming increasingly common.The development of various therapeutic and diagnostic candidates based on the key role of CAMs, withspecial emphasis on integrins in various diseases, as well as the structure–function aspects of cell adhesionand signaling of the different CAMs and ECM are highlighted.

Key words: Integrins, selectins, immunoglobulins, CAM inhibitors, extracellular matrix proteins,αIIbβ3, αvβ3, αvβ5, α4β1, α4β7, α5β1, ICAM, VCAM, PECAM, soluble adhesion molecules,angiogenesis, apopotosis, thrombosis, restenosis, osteoporosis, inflammatory and immunedisorders.

1. Introduction

Many physiological processes, including cell activation, migration,proliferation, and differentiation require direct contact betweencells or cells and extracellular matrix (ECM) proteins. Cell–celland cell–matrix interactions are mediated through several differ-ent families of cell adhesion molecules (CAMs), which includeselectins, integrins, cadherins, and immunoglobulins. The discov-ery of new CAMs, along with new roles for integrins, selectins,


261

262 Mousa

and immunoglobulins in certain disease states, provides greatopportunities for the development of novel therapeutic and per-haps diagnostic modalities.

Intensified drug discovery efforts directed at manipulatingCAM activity through monoclonal antibodies, peptides, pep-tidomimetics, and non-peptide small molecules for diagnostic andtherapeutic applications continue to broaden the scope of keyclinical approaches. This chapter focuses on current advances inthe discovery and development of novel anti-integrins for poten-tial therapeutic and diagnostic applications, as well key methodsof studying the different CAMs.

CAMs play a very significant and critical role in both nor-mal and pathophysiological disease states. For this key reason,the selection of specific and relevant CAMs to target certaindisease conditions without interfering with other normal cellu-lar functions is an important prerequisite for the ultimate suc-cess of an active and safe therapeutic strategy (1, 2). Excitingadvances in our understanding of several CAMs, notably αvβ3,αvβ5, α4β1, α5β1, and αIIb/β3 integrin receptors and theirdirect relationships to different disease states open the door totremendous therapeutic and diagnostic opportunities (1–7). Spe-cific CAMs have been implicated in several different disease states,including cardiovascular disease and cancer, and are involved in anumber of biological systems, including the inflammatory, ocular,pulmonary, bone, central nervous system, kidney, and gastroin-testinal systems. The role of integrin αIIbβ3 in the prevention,treatment and diagnosis of various thromboembolic disorders isexcellent proof of this concept (3–7). In addition, the poten-tial prophylactic role of anti-selectins, the role of β1 and otherleukointegrins in various inflammatory conditions, the potentialutility of soluble adhesion molecules as surrogate markers foracute and chronic endothelial injury, and the potential role ofαvβ3 in angiogenesis and osteoporosis have been reported (8, 9).

The following sections will describe in more detail the thera-peutic and diagnostic applications of selected CAMs.

2. Selectinsand Anti-selectins

Selectins comprise a family of three types of cell adhesion recep-tors (E-, L-, and P-selectins) that share common structuralfeatures, namely a lectin (L), EGF-like (E), and complement(C) binding-like domain (also termed LEC-CAMs). Function-ally, selectins all mediate cellular interactions through the lectindomain of the selectin and cell surface carbohydrate ligands (10).P- and E-selections are calcium-dependent and are expressed

Adhesion Molecules: Potential Therapeutic and Diagnostic Implications 263

on platelets or endothelial cells, where they mediate leuko-cyte adhesion through recognition of cell-specific carbohydrateligands. L-selectins are found on all leukocytes and bind to theircognate receptor, Gly-CAM-1, a mucin-like endothelial glycopro-tein (GP), on endothelial cells (11). E-selectin is an endothelialadhesion molecule whose expression is induced by various inflam-matory stimuli. P-selectin is stored in alpha granules of platelets,as well as Weible-Palade bodies of endothelial cells. E-selectin rec-ognizes the cell surface carbohydrate sialyl lewis x (SLeX) (12),while P-selectin recognizes a carbohydrate moiety that is closelyrelated to SLeX (13).

The selectin family of CAMs plays a key role in early neu-trophil (PMN) rolling and adherence to endothelial cells (ECs).P-selectin on platelet and EC surfaces acts in concert withL-selectin on the leukocyte surface to promote PMN–EC andPMN–platelet interactions. Neutralizing monoclonal antibod-ies directed against P- or L-selectin preserve endothelial andmonocyte cell function in an experimental model of myocardialischemia/reperfusion injury (14, 15). In P-selectin deficient mice,P-selectin has been shown to play a role in neointima formationand potentially impact restenosis (16). L-selectin has also beenshown to mediate PMN rolling interactions at sites of inflamma-tion (17).

2.1. Assay for HumanSoluble Selectin (sP-,sL-or sE-selectin)

The mostly commonly used assay is a quantitative sandwichimmunoassay technique. A microplate is pre-coated with a mon-oclonal antibody specific for sP-, sL- or sE-selectin. Standards,samples, and controls are dispensed into individual wells, togetherwith a horseradish peroxidase (HRP)-conjugated polyclonal anti-body specific for sP-, sL- or sE-selectin. After removal of unboundHRP-conjugated secondary antibody, a substrate is added andcolor develops proportional to selectin concentration.

2.1.1. Procedure 1. Dilute samples as follows: for serum or plasma, a dilutionof 1:100 should be adequate; for cell culture supernatants,use a dilution of 1:25.

2. Remove unused microtiter plates from the frame in whichthey were supplied, and store in a sealed foil pouch with asilica gel sachet.

3. Add 100 μL of standard, diluted sample or diluted param-eter control to wells in duplicate.

4. Cover the plate with a plate sealer and incubate at roomtemperature for 1 h.

5. Add 100 μL of HRP-conjugated anti-selectin to each wellwith sufficient force to ensure mixing. Conjugate is red col-ored to facilitate accuracy in dispensing.

264 Mousa

6. Cover the plate with a new plate sealer (provided bythe manufacturer) and incubate at room temperature for30 min.

7. Aspirate or decant contents from each well and wash with400 μl of wash buffer per well. Repeat the process fivetimes for a total of six washes. After the last wash, aspi-rate or decant the contents and remove any remaining washbuffer by tapping the inverted plate firmly on a clean papertowel.

8. Add 100 μl of substrate to each well. Cover the plate andincubate at room temperature for 30 min.

9. Add 100 μl of stop solution to each well. The stop solu-tion should be added to the wells in the same order as thesubstrate.

10. Determine the optical density (OD) of each well within30 min using a microtiter plate reader (photometer) set at450 nm with a correction wavelength of 620 nm. If thewavelength correction facility is not available, scan plates at450 nm and then separately at 620 nm. Subtract OD620from OD450.

3. Integrins

Integrins are a widely expressed family of cell adhesion receptorsthrough which cells attach to the ECM, to each other, or to het-erologous cells. All integrins are heterodimers composed of an α

and β subunit. They are expressed on a wide variety of cells, andmost cells express several different integrins. The interaction ofintegrins with the cytoskeleton and ECM appears to require thepresence of both subunits. The binding of integrins to their lig-ands is cation dependent. Integrins recognize specific amino acidsequence motifs, the most well characterized of which is the RGDsequence found within a number of matrix proteins, includingfibrinogen, vitronectin, fibronectin, thrombospondin, osteopon-tin, von Willebrand factor (vWF), and others. Integrins also bindto ligands via non-RGD sequences, such as the LDV sequencewithin the CS-1 region of fibronectin recognized by α4β1 inte-grin receptors. There are at least 8 known β subunits and 14 α

subunits (1, 2).Integrin receptors contain an extracellular domain that

engages adhesive ligands and a cytoplasmic face that engagesintracellular proteins. Both of these interactions are critical for celladhesion and anchorage-dependent signal transduction in normaland pathological states. For example, platelet activation induces a


confirmational change in integrin αIIb/β3, thereby converting itinto a high affinity fibrinogen receptor. Fibrinogen binding trig-gers a cascade of protein tyrosine phosphorylation and dephos-phorylation through specific kinases and phosphatases, result-ing in the recruitment of numerous other signaling moleculesinto macromolecular F-actin-rich cytoskeletal complexes assem-bled proximal to the cytoplasmic tails of the αIIb and β3 subunits.These dynamic structures regulate platelet function by coordi-nating and integrating signals emanating from integrins and Gprotein-linked receptors. Studies of integrin mutations confirmthat the cytoplasmic tail of integrin αIIbβ3 is involved in signaltransduction through direct interactions with cytoskeletal and sig-naling molecules. In terms of clinical relevance, blocking fibrino-gen binding to the extracellular domain of integrin aIIbβ3 hasbeen shown to effectively prevent the formation of platelet-richarterial thrombi after coronary angioplasty (18). Once the fullcohort of proteins that interact with the cytoplasmic tails of inte-grin αIIbβ3 are identified, it may be possible to develop selectiveinhibitors of integrin adhesion or signaling whose sites of actionare intracellular.

The commercial and therapeutic potential of CAMs is on therise. The discovery of new CAMs, along with new roles for inte-grins, selectins, and immunoglobulins in certain disease states,opens the door to important opportunities in the developmentof therapeutic and diagnostic agents. Integrins represent one ofthe best opportunities for achieving small molecule antagonistsfor both therapeutic and diagnostic applications in several impor-tant diseases with unmet medical needs.

3.1. Potentand SelectiveSmall-MoleculeAntagonistsof α4 Integrins

The α4 integrins are central to leukocyte–cell and leukocyte–matrix adhesive interactions. Integrin α4β7 expression is restrictedto leukocytes, with the exception of neutrophils. It interactswith the immunoglobulin superfamily member vascular-cell adhe-sion molecule 1 (VCAM-1), and an alternately spliced formof fibronectin (FN), as well as mucosal addressin-cell adhesionmolecule 1 (MAdCAM-1), a mucosal vascular addressin, or hom-ing receptor, that contains immunoglobulin-like domains relatedto VCAM-1 (19). Monoclonal antibodies to the α4 subunit orα4β7 can block the adhesive function of α4 and/or α4β7 integrinsin vitro. Studies in vivo with these monoclonal antibodies in sev-eral species demonstrate that the interactions mediated by α4 inte-grins play key pathophysiological roles in immune and inflamma-tory reactions. Thus, α4 integrin-dependent adhesive interactionswith VCAM-1, MAdCAM, and FN appear to play a central rolein the recruitment, priming, activation, and apoptosis of certainleukocyte subsets. As such, α4 integrins represent novel targetsfor drug intervention. A selective and potent anti-α4 monoclonalantibody and small-molecule antagonists have shown in vivo

266 Mousa

efficacy in several experimental animal models (20, 21).MAdCAM-1 is an immunoglobulin-like adhesion receptor that ispreferentially expressed by venular endothelial cells, defining sitesof lymphocyte extravasation in mucosal lymphoid tissues and lam-ina popria. Peptide-based analogs based on various regions in thefirst and second immunoglobulin domains of MAdCAM-1 thatmimic the binding of α4β7 have been identified.

3.2. β1 Integrins The β1 integrins, also known as the very late activation anti-gen (VLA) subfamily due to the late appearance of VLA afteractivation, comprise the largest subfamily of integrins. There areat least seven receptors of this subfamily, each with a differ-ent ligand specificity. Among the most well characterized areα4β1, α5β1, α6β1, and αnβ1 receptors. The leukocyte inte-grin α4β1 (also known as VLA-4 and CD49d/CD29) is acell adhesion receptor expressed predominantly on lymphocytes,monocytes, and eosinophils (22). VLA-4 has been suggestedas a potential target for therapeutics in chronic inflammatorydiseases.

3.2.1. Leukocyte Integrinα4β1 as a PotentialTherapeutic Target

Leukocyte populations that express α4β1 primarily mediatechronic inflammatory diseases (i.e., rheumatoid arthritis, asthma,psoriasis, and allergy). In contrast, VLA-4 is not present on cir-culating unstimulated neutrophils, which constitute a first lineof defense against acute infections. Eosinophils selectively accu-mulate at sites of pulmonary inflammation in chronic aller-gic diseases such as bronchial asthma. The role of β1 inte-grins and their regulation by cytokines and other inflamma-tory mediators during eosinophil adhesion to the endothe-lium and ECM and during transendothelial migration has beenwell documented (23, 24). Interactions of VLA-4 with alterna-tively spliced forms of fibronectin containing the CS-1 regionhave been exploited in the design of small molecule inhibitorsthat bind to VLA-4 and block receptor function. Evaluationof these analogs in animal models of disease indicates thatVLA-4 receptor blockers have the potential to achieve dramaticin vivo results in a variety of chronic inflammatory disorders(20–22).

Infiltration of circulating immune cells into the central ner-vous system (CNS) can result in edema, myelin damage, andparalysis (25). Importantly, a role for integrin α4β1 in thisinfiltration process has been demonstrated. When administeredto animals with experimental autoimmune encephalomyelitis(EAE), antibodies against α4 integrin prevent the adhesion oflymphocytes and monocytes to inflamed endothelia within bloodvessels of the CNS and prevent immune cell infiltration. Evenwhen administered to animals after the onset of paralysis, anti-α4


antibodies reversed all clinical signs of disease. Magnetic res-onance imaging analysis of these animals showed that anti-body treatment reduced edema and reduced the permeabilityof the blood brain barrier to gadnolium-DPTA, and histolog-ical analysis demonstrated that antibody treatment preventedthe destruction of myelin. Remarkably, anti-α4 antibodiesreversed the accumulation of lymphocytes and monocytes withinthe CNS while having no affect on the level of these cellsin circulation. These results suggest that the active diseaseprocess in EAE requires ongoing recruitment of circulat-ing cells into the CNS and that antibodies directed againstα4 integrins prevent this recruitment and reverse diseaseprogression.

3.2.2. β1 Integrins inGastrointestinal (GI)Disease

Inflammatory bowel disease (IBD), Crohn’s disease, and ulcera-tive colitis (UC) are immunologically-based illnesses. The patternof expression of β integrins in isolated intestinal lamina mononu-clear cells from IBD and normal intestines is similar to that of nor-mal solid organs. Isolated CD3+ cells from patients with Crohn’sdisease express more β1 than normal individuals, supporting theidea that there are distinct β integrin systems involved in GIdiseases. Interest in β1 and β7 integrins in particular as poten-tial therapeutic targets for GI inflammatory disease remains high(26).

3.2.3. α5β1 Integrinin Angiogenesisand Bacterial Infection

Recent evidence suggests that α5β1 integrin is involved in themodulation of angiogenesis (27), suggesting that α5β1 antago-nists might be useful in various angiogenesis-mediated disorders(27). Similarly, α5β1 integrin has been implicated in mediatingbacterial invasion into human host cells leading to antibiotic resis-tance (28).

3.3. β2 Integrins The leukocyte-restricted β2 (CD18) integrins promote a vari-ety of homotypic and heterotypic cell adhesion events requiredfor normal and pathologic functioning of the immune sys-tem (2). Several physiological processes, including cell adhe-sion, activation, migration, and transmigration, require directcontact between cells or ECM proteins via CAM receptors. Todate, three members of this integrin subfamily have been iden-tified: CD11a/CD18 (LFA-1), CD11b/CD18 (Mac-1), andCD11c/CD18 (P150,95). A fourth α chain, designated αd, hasbeen cloned and shown to associate with CD18 in normal leuko-cytes and upon co-transfection into CHO cells. In vitro studieshave demonstrated that LFA-1 and Mac-1 on neutrophils canbe differentially activated, with distinct functional consequences(2). Studies in CD11b-deficient mice further underscore the bio-logic significance of distinct contributions of LFA-1 and Mac-1to neutrophil-dependent tissue injury.

268 Mousa

3.4. αIIb/β3 Integrin

3.4.1. IntravenousPlatelet αIIb/β3Receptor Antagonists –Potential Clinical Utility

The realization that platelet integrin αIIbβ3 is the final com-mon pathway for platelet aggregation regardless of the activa-tion mechanism has prompted the development of several small-molecule αIIbβ3 receptor antagonists for intravenous and/or oralanti-thrombotic use. Platelet αIIbβ3 receptor blockade representsa very promising therapeutic and diagnostic strategy for treat-ing thromboembolic disorders. Clinical experience (i.e., stud-ies of efficacy/safety) gained with injectable αIIbβ3 antagonistspromises to yield valuable insight into the potential of long-term chronic use of oral αIIb/β3 antagonists. At this point,there are still many unanswered questions, however, and care-ful study will be needed to determine the safety and efficacyof this approach, either alone or in combination with otherantiplatelet/anticoagulant therapies.

Abciximab. The clinical utility of abciximab (ReoPro, c7E3Fab) has been demonstrated in several trials involving coronaryartery intervention procedures (29–31). The potent, rapid, andsustained block of platelet GPIIb/IIIa receptors, and perhapsαvβ3 as well, might be a key component of abciximab’s abilityto mediate dramatic early anti-thrombotic benefits. Early bene-fits were maintained for over 3 years in patients receiving 12-habciximab treatment in the EPIC (Evaluation of 7E3 for the Pre-vention of Ischemic Complications) trial.

Integrilin. The IMPACT II (coronary intervention; broadentry criteria) and PURSUIT (unstable angina; chest pain <24 h,ischemic ECG changes) trials both demonstrated significant clin-ical benefits of integrilin, a cyclic heptapeptide KGD analog (32).

Tirofiban. The RESTORE (coronary intervention; high riskof abrupt closure as per clinical and anatomic criteria), PRISM,and PRISM-plus (unstable angina; chest pain at 24 and 12 h)trials have demonstrated significant clinical benefits of tirofiban.

Lamifiban. The PARAGON trial demonstrated significantclinical benefits of lamifiban (33). However, studies with lami-fiban in Canada were stopped due to lack of efficacy and nuisancebleeding.

3.4.2. Chronic Therapywith Platelet GPIIb/IIIaAntagonists

When used for acute therapy of coronary arterial disease,GPIIb/IIIa antagonists must exhibit a high degree of plateletantagonism. By comparison, the requirements for chronic ther-apy using orally active agents have only recently been investi-gated. Interactions of oral GPIIb/IIIa antagonists with aspirinand other antiplatelet and anticoagulant drugs lead to shifts inthe dose response curves for both efficacy and unwanted sideeffects, such as increased bleeding time (34–36). More recently,xemilofiban (EXCITE trial) and orbofiban (OUPIS trial), spon-


sored by Searle, and sibrafiban, sponsored by Roche, were with-drawn because of disappointing outcomes. These results raise seri-ous questions about the potential of oral GPIIb/IIIa antagonists(37, 38).

Issues associated with oral GPIIb/IIIa antagonists currentlyunder clinical investigation include thrombocytopenia, monitor-ing, bleeding risk, and drug interactions (39, 40). The role of theplatelet integrin GPIIb/IIIa receptor and its potential utility as aradiodiagnostic agent for the rapid detection and/or diagnosis ofthromboembolic events has been demonstrated (41).

3.5. αvβ3 Integrinand Matrix Proteinsin VascularRemodeling

Processes involved in vascular remodeling also play a key role inthe pathological mechanisms of atherosclerosis and restenosis. Inresponse to vascular injury induced by percutaneous translumi-nal angioplasty (PTA), matrix proteins like osteopontin and vit-ronectin are rapidly up-regulated (42). Osteopontin stimulatessmooth muscle cell migration through its action on integrin αvβ3and thereby contributes to neointima formation and resteno-sis (43, 44). Osteopontin and vitronectin induce angiogenesis,which may also support neointima formation and arteriosclerosis(45). Thus, specific matrix proteins acting through integrin recep-tors, in particular, αvβ3, represent important targets for selec-tive antagonists aimed at blocking the pathological processes ofrestenosis (42). Integrin αvβ3 ligands have also been utilized forthe site-directed delivery of different therapeutic and diagnos-tic targets in oncology, in addition to having anti-cancer effects(46, 47).

4. Cellular andIntegrin-BasedAntiplateletEfficacy Assays

4.1. LightTransmittanceAggregometry Assay

Venous blood is obtained from healthy non-smoker and non-fasted human donors (35–45 years of age, males and females)who have been drug- and aspirin-free for at least 2 weeks priorto blood collection (5, 6). Briefly, blood is collected into cit-rated Vacutainer tubes. The blood is subjected to centrifugationfor 10 min at 150×g in a Sorvall RT6000 tabletop centrifugeequipped with a H-1000 B rotor at room temperature, afterwhich platelet-rich plasma (PRP) is removed. After centrifuga-tion of the remaining blood for 10 minutes at 1,500×g at roomtemperature, platelet-poor plasma (PPP) is removed. Samples areassayed on a PAP-4 Platelet Profiler, using PPP as the blank (set as100% transmittance). PRP (200 μL at a concentration of 2×108

platelets/ml) is added to individual microcentrifuge tubes, and

270 Mousa

transmittance is set to 0%. The platelet agonist ADP (20 μL,10 μM final concentration) is added to each tube and aggrega-tion profiles are plotted as % transmittance over time. Antiplateletagent (20 μl) is added at different concentrations ranging from0.001–100 μM for 8 min prior to the addition of ADP. Resultsare expressed as percent inhibition of agonist-induced plateletaggregation, or IC50 (μM).

4.2. Platelet125I-FibrinogenBinding Assay

Human PRP (h-PRP) is subjected to size exclusion chromatogra-phy to prepare human-gel-purified platelets (h-GPP). Aliquots ofh-GPP (2×108 platelets/ml) along with 1 mM calcium chlorideare added to removable 96-well plates. 125I-Fibrinogen (26.5μCi/mg) is added, and h-GPP are activated by the addition ofADP, epinephrine, and sodium arachidonate (100 μM each).125I-Fibrinogen bound to activated platelets is separated from thefree form by centrifugation, and the radioactivity of the samplesis counted using a gamma counter. Non-specific binding (due toentrapment of 125I-fibrinogen) in the presence or absence of testagent (and in the absence of agonists) should be in the range of4–6% of total 125I-fibrinogen binding to agonist-activatedplatelets.

4.3. 293 β FibrinogenAdhesion Assay

In this assay, αvβ3-transfected 293 cells are used. Adhesion of293/β3 cells to fibrinogen is completely inhibited by an anti-αvβ3monoclonal antibody, indicating that fibrinogen binding is depen-dent on integrin αvβ3. ELISA plates are coated with fibrinogenat a concentration of 25 μg/well and stored at 4ºC until use.On the day of the assay, plates are washed twice with phosphate-buffered saline (PBS) without cations, and the wells are incubatedwith 5% BSA/PBS for 2 h. 293/β3 cells at 30–70% confluenceare harvested and resuspended at a density of 1×106 cells/ml. Toeach well of the 96-well plate, 65 μl of buffer is added, followedby 5 μl of different concentrations of test agent. Cells (130 μl)are added and the plates are incubated at 37ºC in 5% CO2 for15 min. Non-adherent cells are removed, and remaining cells arelysed in a solution of 100 mM potassium phosphate, 0.2% TritonX-100, pH 7.8. An aliquot (5 ul) is assayed for β-galactosidaseusing a standard luminescence assay. Luminescence values areconverted to β-galactosidase units using a standard curve, anddata are normalized to correct for non-specific activity. Data arepresented as percent inhibition and/or IC50 values.

4.4. SK-BR-3Cell-Vitronectin(αvβ5-Mediated)Adhesion Assay

An αvβ5-expressing breast cancer cell line (SK-BR-3; ATCC,Rockville, MD) is used for this assay. Adhesion of SK-BR-3 cellsto vitronectin is αvβ5-dependent, as evidenced by complete inhi-bition of binding by an anti-αvβ5 monoclonal antibody (24). ACostar multi-well tissue culture plate is coated with 100 μl ofvitronectin (0.25 μg per well) overnight at 4◦C. The plate is


washed twice with 200 μl of PBS, and then non-specific binding isblocked by incubating the plates with 200 μl of PBS+5% BSA for1 h at room temperature. Cells are labeled with 2 μM calcein-AM(Molecular Probes) for 30 min at 37◦C in a humidified incuba-tor. Cells (1×106 cells/ml) are pre-incubated with 150 μl of testcompound, or medium as a control, gently mixed, and then incu-bated for 15 min at room temperature. Treated cells are added tothe assay plate in duplicate and then incubated for 60 min withshaking at room temperature. Plates are covered with foil to pre-vent photobleaching. Plates are washed to remove non-adherentcells, 100 μl of media is added to each well, and then fluorescenceis measured using a Cytofluor 2300 system (excitation, 485 nM;emission, 530 nm).

4.5. Purified α5β1Receptor-BiotinylatedFibronectin BindingAssay

Purified integrin α5β1 obtained from human placenta is coatedonto Costar high capacity binding plates overnight at 4◦C. Thecoating solution is discarded and plates are washed once withbuffer. Wells are incubated with 200 μL of buffer containing1% BSA. After washing once with buffer, 100 μl of biotinylatedfibronectin (2 nM), plus 11 μl of test agent or buffer/BSA isadded to each well and then the plates are incubated for 1 h atroom temperature. Plates are washed twice with buffer and thenincubated for 1 h at room temperature with 100 μl of alkalinephosphatase-conjugated anti-biotin antibody. Plates are washedtwice with buffer and then incubated for 1 h with 100 μl of alka-line phosphatase substrate. Color is developed at room tempera-ture for approximately 45 min, and then the reaction is stoppedby the addition of 2 N NaOH. Plates are read at 405 nm.

Functional assays of integrin-mediated intracellular signalingcan be performed using transfected cells or other cell systems thatexpress specific integrins and neutralizing or blocking antibod-ies or small molecule ligands specific for the integrin of inter-est. Integrin-mediated cell migration and/or proliferation can bestudied using classical migration or proliferation assays.

5. Immunoglobulins

Inter-cellular cell adhesion molecule (ICAM) and VCAM aremembers of the immunoglobulin (Ig) superfamily. At present, agreat deal of effort in the targeting of Ig superfamily membersis focused on the development of specific monoclonal antibod-ies and/or anti-sense oligonucleotides and small molecules thatspecifically block transcription factors. Strategies for the designof direct small molecular weight inhibitors of the Ig superfamilyare somewhat more problematic. However, current advances in

272 Mousa

molecular modeling combined with advances in crystal structuredata open the door to the development of cyclic peptides andpeptidomimetic Ig antagonists.

Monoclonal antibodies to ICAM-1 have anti-inflammatoryproperties, and show tremendous therapeutic potential in liverand kidney transplants, as well as in rheumatoid arthritis (48, 49).In contrast to current immunosuppressants, which are efficaciousin organ transplant but have major adverse effects, anti-CAMsmay prove to be more effective and safer.

5.1. Role of PECAM-1in RegulatingTransendothelialMigration of PMNsin Disease States

PMN’s adhere to the inflamed vascular endothelium, even-tually undergoing transendothelial migration. The transmigra-tion process is regulated largely by platelet/endothelial celladhesion molecule-1 (PECAM-1), an adhesion receptor thatis expressed on platelets, leukocytes and at the intercellularjunctions of endothelial cells. PECAM-1 neutralizing antibodiesselectively block PMN migration and markedly attenuate injuryto ischemic-reperfused myocardium and coronary endothelium.Intravital microscopy demonstrated that the protective mech-anism of PECAM-1 blockade involves inhibition of PMNtransendothelial migration (50). Anti-PECAM-1 has been widelyused as an indicator, typically in conjunction with immunostain-ing, of endothelial cells in various histological and molecularstudies (51).

5.2. Soluble AdhesionMolecules asSurrogate Markers

CAMs are well recognized as adhesive receptors that facilitateadhesion, migration and transmigration of circulating cells intodamaged vascular tissues. Recent studies have demonstrated thatICAM-1 is expressed on human athersclerotic plaques, and thattreatment with an anti-ICAM-1 monoclonal antibody results ina significant reduction of myocardial infarct size in experimentalmyocardial/ischemia reperfusion injury models (55, 56). In addi-tion, soluble isoforms of CAMs are believed to be shed from thesurface of activated cells, and can be quantified in peripheral blood(52, 53). Increased serum concentrations of soluble CAMs havebeen documented in a variety of diseases (52, 53), suggesting theprognostic and diagnostic potential of various soluble adhesionmolecules in vascular and cardiovascular disease.

5.3. Human Soluble(s)VCAM-1, ICAM-1,or PECAM Assay

The assay is based on the simultaneous binding of sVCAM-1present in the sample or standard to two antibodies directedagainst different epitopes on the sVCAM-1 molecule. One anti-body is adsorbed onto the surface of a microtiter well, andthe other is conjugated to HRP. Any sVCAM-1 (or sICAM-1 or PECAM) present will bridge the two different antibodies,allowing detection of the entire complex (antibody-sVCAM-1-antibody). Briefly, after removal of unbound material by aspi-ration and washing, the amount of bound antibody complex


is detected by reaction with an HRP substrate, which yields acolored product proportional to the amount of complex (andthus sVCAM-1) in the sample. The colored product can be quan-tified photometrically. By analyzing standards of known sVCAM-1 concentration at the same time as the test samples, the concen-tration of the samples can be determined on a plot of signal versusconcentration.

5.3.1. Procedure For most samples (serum, plasma, or cell culture supernatant), adilution of 1:50 should be adequate.

1. Add 100 μl of diluted anti-VCAM-1-HRP conjugate to eachwell.

2. Add 100 μl of standard, diluted sample, or diluted parametercontrol to each well with sufficient force to ensure mixing.Shaking or tapping is not recommended.

3. Cover the plate with a plate sealer and incubate at room tem-perature for 1.5 h.

4. Aspirate or decant contents from each well and wash byadding 300 μl of wash buffer per well. Repeat the processfive times for a total of six washes.

5. After the last wash, aspirate or decant the contents andremove any remaining wash buffer by tapping the invertedplate firmly on clean paper toweling.

6. Immediately after decanting, add 100 μl of substrate to eachwell.

7. Cover the plate with a plate sealer and incubate at room tem-perature for 30 min.

8. Add 100 μl of stop solution to each well. The stop solu-tion should be added to the wells in the same order as thesubstrate.

9. Determine the OD of each well within 30 min using amicrotiter plate reader or Photometer set at 450 nm with acorrection wavelength of 620 nm. The photometer or platereader should be blanked according to the manufacturer’sinstructions. If the wavelength correction function is notavailable, read plates at 450 nm and then again at 620 nm.Subtract OD620 from the OD450.

6. Conclusion

Several members of the different CAM superfamilies, in particularthe integrin family, represent excellent potential therapeutic anddiagnostic targets for a number of diseases with unmet medical

274 Mousa

needs. The identity of the CAM within a specific pathophysiolog-ical aspect of the disease process as well as the ease in achievinga small molecule anti-CAM candidate will ultimately determinethe success of this particular class of therapeutic and diagnosticagent.

References

1. Cox, D., Aoki, T., Seki, J., Motoyama, Y.,and Yoshida, K. (1994) The pharmacology ofthe integrins Med Res Rev 14, 195–228.

2. Lal, H., Verma, S.K., Foster, D.M., Golden,H.B., Reneau, J.C., Watson, L.E., Singh, H.,and Dostal, D.E. (2009) Integrins and prox-imal signaling mechanisms in cardiovasculardisease Front Biosci 14, 2307–34.

3. Gold, H.K., Gimple, L.W., Yasuda, T., Lein-bach, R.C., Werner, W., Holt, R., Jordan,R., Berger, H., Collen, D., and Coller, B.S.(1990) Pharmacodynamic study of F(ab’)2fragments of murine monoclonal antibody7E3 directed against human platelet glyco-protein IIb/IIIa in patients with unstableangina pectoris J Clin Invest 86, 651–9.

4. Mousa, S., and Topol, E. Novel antiplatelettherapies: recent advances in the develop-ment of platelet GPIIb/IIIa receptor antag-onists. In: Serruys, P.W., and Holmes, D.,eds. Current Review of Interventional Cardi-ology, 3rd edition. Philadelphia, PA: CurrentMedicine;1997:114–29.

5. Mousa, S.A., Bozarth, J.M., Forsythe, M.S.,Jackson, S.M., Leamy, A., Diemer, M.M.,Kapil, R.P., Knabb, R.M., Mayo, M.C.,Pierce, S.K., et al. (1994) Antiplatelet andantithrombotic efficacy of DMP 728, a novelplatelet GPIIb/IIIa receptor antagonist Cir-culation 89, 3–12.

6. Mousa, S.A., Bozarth, J.M., Forsythe, M.S.,Lorelli, W., Thoolen, M.J., Ramachan-dran, N., Jackson, S., De Grado, W., andReilly, T.M. (1993) Antiplatelet efficacy andspecificity of DMP728, a novel plateletGPIIb/IIIa receptor antagonist Cardiology83, 374–82.

7. van’t Hof, A.W., and Valgimigli, M. (2009)Defining the role of platelet glycoproteinreceptor inhibitors in STEMI: focus ontirofiban Drugs 69, 85–100.

8. Brooks, P.C., Clark, R.A., and Cheresh,D.A. (1994) Requirement of vascular inte-grin alpha v beta 3 for angiogenesis Science264, 569–71.

9. Brooks, P.C., Montgomery, A.M., Rosenfeld,M., Reisfeld, R.A., Hu, T., Klier, G., andCheresh, D.A. (1994) Integrin alpha v beta3 antagonists promote tumor regression by

inducing apoptosis of angiogenic blood ves-sels Cell 79, 1157–64.

10. Brandley, B.K., Swiedler, S.J., and Robbins,P.W. (1990) Carbohydrate ligands of theLEC cell adhesion molecules Cell 63,861–3.

11. Lasky, L.A., Singer, M.S., Dowbenko, D.,Imai, Y., Henzel, W.J., Grimley, C., Fennie,C., Gillett, N., Watson, S.R., and Rosen, S.D.(1992) An endothelial ligand for L-selectinis a novel mucin-like molecule Cell 69,927–38.

12. Lasky, L.A. (1992) Selectins: interpreters ofcell-specific carbohydrate information duringinflammation Science 258, 964–9.

13. Phillips, M.L., Nudelman, E., Gaeta, F.C.,Perez, M., Singhal, A.K., Hakomori, S., andPaulson, J.C. (1990) ELAM-1 mediates celladhesion by recognition of a carbohydrateligand, sialyl-Lex Science 250, 1130–2.

14. Mulligan, M.S., Paulson, J.C., De Frees, S.,Zheng, Z.L., Lowe, J.B., and Ward, P.A.(1993) Protective effects of oligosaccharidesin P-selectin-dependent lung injury Nature364, 149–51.

15. Weyrich, A.S., Ma, X.Y., Lefer, D.J., Alber-tine, K.H., and Lefer, A.M. (1993) Invivo neutralization of P-selectin protectsfeline heart and endothelium in myocardialischemia and reperfusion injury J Clin Invest91, 2620–9.

16. Hullinger, T., DeGraaf, G., Hartman, J., andShebuski, R. (1995) The effect of P-selectinblockade on neointimal lesion developmentin a primate carotid injury model FASEB 9.

17. Taylor, M.E., and Drickamer, K. (2007)Paradigms for glycan-binding receptors incell adhesion Curr Opin Cell Biol 19,572–7.

18. Topol, E.J., Califf, R.M., Weisman, H.F.,Ellis, S.G., Tcheng, J.E., Worley, S., Ivanhoe,R., George, B.S., Fintel, D., Weston, M.,et al. (1994) Randomised trial of coronaryintervention with antibody against plateletIIb/IIIa integrin for reduction of clinicalrestenosis: results at six months. The EPICinvestigators Lancet 343, 881–6.

19. Berlin, C., Berg, E.L., Briskin, M.J., Andrew,D.P., Kilshaw, P.J., Holzmann, B., Weissman,


I.L., Hamann, A., and Butcher, E.C. (1993)Alpha 4 beta 7 integrin mediates lymphocytebinding to the mucosal vascular addressinMAdCAM-1 Cell 74, 185–95.

20. Elices, M.J., Osborn, L., Takada, Y., Crouse,C., Luhowskyj, S., Hemler, M.E., andLobb, R.R. (1990) VCAM-1 on activatedendothelium interacts with the leukocyteintegrin VLA-4 at a site distinct from theVLA-4/fibronectin binding site Cell 60,577–84.

21. Issekutz, T.B. (1991) Inhibition of in vivolymphocyte migration to inflammation andhoming to lymphoid tissues by the TA-2monoclonal antibody. A likely role for VLA-4in vivo J Immunol 147, 4178–84.

22. Hamann, A., Andrew, D.P., Jablonski-Westrich, D., Holzmann, B., and Butcher,E.C. (1994) Role of alpha 4-integrins in lym-phocyte homing to mucosal tissues in vivo JImmunol 152, 3282–93.

23. Springer, T.A. (1994) Traffic signals for lym-phocyte recirculation and leukocyte emi-gration: the multistep paradigm Cell 76,301–14.

24. Yednock, T.A., Cannon, C., Fritz, L.C.,Sanchez-Madrid, F., Steinman, L., andKarin, N. (1992) Prevention of experimentalautoimmune encephalomyelitis by antibodiesagainst alpha 4 beta 1 integrin Nature 356,63–6.

25. Cannella, B., and Raine, C.S. (1995) Theadhesion molecule and cytokine profile ofmultiple sclerosis lesions Ann Neurol 37,424–35.

26. Cerf-Bensussan, N., Jarry, A., Brousse,N., Lisowska-Grospierre, B., Guy-Grand,D., and Griscelli, C. (1987) A mono-clonal antibody (HML-1) defining a novelmembrane molecule present on humanintestinal lymphocytes Eur J Immunol 17,1279–85.

27. Kim, S., Bell, K., Mousa, S.A., andVarner, J.A. (2000) Regulation of angio-genesis in vivo by ligation of integrinalpha5beta1 with the central cell-bindingdomain of fibronectin Am J Pathol 156,1345–62.

28. Cue, D., Southern, S.O., Southern, P.J.,Prabhakar, J., Lorelli, W., Smallheer, J.M.,Mousa, S.A., and Cleary, P.P. (2000) Anonpeptide integrin antagonist can inhibitepithelial cell ingestion of Streptococcuspyogenes by blocking formation of inte-grin alpha 5beta 1-fibronectin-M1 proteincomplexes Proc Natl Acad Sci USA 97,2858–63.

29. The EPIC Investigators (1994) Use of amonoclonal antibody directed against the

platelet glycoprotein IIb/IIIa receptor inhigh-risk coronary angioplasty N Engl J Med330, 956–61.

30. Kleiman, N.S., Ohman, E.M., Califf, R.M.,George, B.S., Kereiakes, D., Aguirre, F.V.,Weisman, H., Schaible, T., and Topol, E.J.(1993) Profound inhibition of platelet aggre-gation with monoclonal antibody 7E3 Fabafter thrombolytic therapy. Results of theThrombolysis and Angioplasty in MyocardialInfarction (TAMI) 8 Pilot Study J Am CollCardiol 22, 381–9.

31. Tcheng, J.E., Ellis, S.G., George, B.S.,Kereiakes, D.J., Kleiman, N.S., Talley, J.D.,Wang, A.L., Weisman, H.F., Califf, R.M.,and Topol, E.J. (1994) Pharmacodynamicsof chimeric glycoprotein IIb/IIIa integrinantiplatelet antibody Fab 7E3 in high-risk coronary angioplasty Circulation 90,1757–64.

32. Tcheng, J.E., Harrington, R.A., Kottke-Marchant, K., Kleiman, N.S., Ellis, S.G.,Kereiakes, D.J., Mick, M.J., Navetta, F.I.,Smith, J.E., Worley, S.J., et al. (1995)Multicenter, randomized, double-blind,placebo-controlled trial of the platelet inte-grin glycoprotein IIb/IIIa blocker Integrelinin elective coronary intervention. IMPACTinvestigators Circulation 91, 2151–7.

33. Peerlinck, K., De Lepeleire, I., Goldberg, M.,Farrell, D., Barrett, J., Hand, E., Panebianco,D., Deckmyn, H., Vermylen, J., and Arnout,J. (1993) MK-383 (L-700,462), a selectivenonpeptide platelet glycoprotein IIb/IIIaantagonist, is active in man Circulation 88,1512–7.

34. Cannon, C.P., McCabe, C.H., Borzak, S.,Henry, T.D., Tischler, M.D., Mueller, H.S.,Feldman, R., Palmeri, S.T., Ault, K., Hamil-ton, S.A., Rothman, J.M., Novotny, W.F.,and Braunwald, E. (1998) Randomized trialof an oral platelet glycoprotein IIb/IIIaantagonist, sibrafiban, in patients after anacute coronary syndrome: results of theTIMI 12 trial. Thrombolysis in myocardialinfarction Circulation 97, 340–9.

35. Muller, T.H., Weisenberger, H., Brickl, R.,Narjes, H., Himmelsbach, F., and Krause,J. (1997) Profound and sustained inhibi-tion of platelet aggregation by fradafiban,a nonpeptide platelet glycoprotein IIb/IIIaantagonist, and its orally active prodrug,lefradafiban, in men Circulation 96, 1130–8.

36. Simpfendorfer, C., Kottke-Marchant, K.,Lowrie, M., Anders, R.J., Burns, D.M.,Miller, D.P., Cove, C.S., DeFranco, A.C.,Ellis, S.G., Moliterno, D.J., Raymond, R.E.,Sutton, J.M., and Topol, E.J. (1997)First chronic platelet glycoprotein IIb/IIIa

276 Mousa

integrin blockade. A randomized, placebo-controlled pilot study of xemilofiban inunstable angina with percutaneous coronaryinterventions Circulation 96, 76–81.

37. Harrington, R.A., Armstrong, P.W.,Graffagnino, C., Van De Werf, F., Kereiakes,D.J., Sigmon, K.N., Card, T., Joseph, D.M.,Samuels, R., Granett, J., Chan, R., Califf,R.M., and Topol, E.J. (2000) Dose-finding,safety, and tolerability study of an oral plateletglycoprotein IIb/IIIa inhibitor, lotrafiban, inpatients with coronary or cerebral atheroscle-rotic disease Circulation 102, 728–35.

38. Mousa, S., and Wityak, J. (1998) Orallyactive Isoxazoline GOIIb/IIIa antagonistsCardiovas Drug Rev 16, 48–61.

39. Quinn, M., and Fitzgerald, D.J. (1998)Long-term administration of glycoproteinIIb/IIIa antagonists Am Heart J 135,S113–8.

40. Vorchheimer, D.A., and Fuster, V. (1998)Oral platelet glycoprotein IIb/IIIa receptorantagonists: the present challenge is safetyCirculation 97, 312–4.

41. Mousa, S.A., Bozarth, J.M., Edwards, S.,Carroll, T., and Barrett, J. (1998) Noveltechnetium-99m-labeled platelet GPIIb/IIIareceptor antagonists as potential imagingagents for venous and arterial thrombosisCoron Artery Dis 9, 131–41.

42. Srivatsa, S.S., Fitzpatrick, L.A., Tsao, P.W.,Reilly, T.M., Holmes, D.R., Jr., Schwartz,R.S., and Mousa, S.A. (1997) Selective alphav beta 3 integrin blockade potently limitsneointimal hyperplasia and lumen stenosisfollowing deep coronary arterial stent injury:evidence for the functional importance ofintegrin alpha v beta 3 and osteopontinexpression during neointima formation Car-diovasc Res 36, 408–28.

43. Liaw, L., Skinner, M.P., Raines, E.W., Ross,R., Cheresh, D.A., Schwartz, S.M., andGiachelli, C.M. (1995) The adhesive andmigratory effects of osteopontin are medi-ated via distinct cell surface integrins. Role ofalpha v beta 3 in smooth muscle cell migra-tion to osteopontin in vitro J Clin Invest 95,713–24.

44. Yue, T.L., McKenna, P.J., Ohlstein, E.H.,Farach-Carson, M.C., Butler, W.T., Johan-son, K., McDevitt, P., Feuerstein, G.Z., andStadel, J.M. (1994) Osteopontin-stimulatedvascular smooth muscle cell migration ismediated by beta 3 integrin Exp Cell Res214, 459–64.

45. van der Zee, R., Murohara, T., Passeri, J.,Kearney, M., Cheresh, D.A., and Isner, J.M.(1998) Reduced intimal thickening follow-ing alpha(v)beta3 blockade is associated withsmooth muscle cell apoptosis Cell AdhesCommun 6, 371–9.

46. Kerr, J.S., Slee, A.M., and Mousa, S.A.(2002) The alpha v integrin antagonists asnovel anticancer agents: an update ExpertOpin Investig Drugs 11, 1765–74.

47. Oba, M., Fukushima, S., Kanayama, N.,Aoyagi, K., Nishiyama, N., Koyama, H., andKataoka, K. (2007) Cyclic RGD peptide-conjugated polyplex micelles as a targetablegene delivery system directed to cells possess-ing alphavbeta3 and alphavbeta5 integrinsBioconjug Chem 18, 1415–23.

48. Flavin, T., Ivens, K., Rothlein, R., Faanes, R.,Clayberger, C., Billingham, M., and Starnes,V.A. (1991) Monoclonal antibodies againstintercellular adhesion molecule 1 prolongcardiac allograft survival in cynomolgus mon-keys Transplant Proc 23, 533–4.

49. Haug, C.E., Colvin, R.B., Delmonico,F.L., Auchincloss, H., Jr., Tolkoff-Rubin,N., Preffer, F.I., Rothlein, R., Norris, S.,Scharschmidt, L., and Cosimi, A.B. (1993)A phase I trial of immunosuppression withanti-ICAM-1 (CD54) mAb in renal allograftrecipients Transplantation 55, 766–72; Dis-cussion 72–3.

50. Rosenblum, W.I., Nelson, G.H., Worm-ley, B., Werner, P., Wang, J., and Shih,C.C. (1996) Role of platelet-endothelialcell adhesion molecule (PECAM) in plateletadhesion/aggregation over injured but notdenuded endothelium in vivo and ex vivoStroke 27, 709–11.

51. Muller, W.A., and Randolph, G.J. (1999)Migration of leukocytes across endotheliumand beyond: molecules involved in the trans-migration and fate of monocytes J Leukoc Biol66, 698–704.

52. Gearing, A.J., and Newman, W. (1993)Circulating adhesion molecules in diseaseImmunol Today 14, 506–12.

53. Newman, W., Beall, L.D., Carson, C.W.,Hunder, G.G., Graben, N., Randhawa,Z.I., Gopal, T.V., Wiener-Kronish, J., andMatthay, M.A. (1993) Soluble E-selectin isfound in supernatants of activated endothe-lial cells and is elevated in the serum ofpatients with septic shock J Immunol 150,644–54.

Chapter 12

Pharmacogenomics in Thrombosis

Shaker A. Mousa

Abstract

Inherited or acquired genetic abnormalities play a major role in thromboembolic complications. Thegoal of pharmacogenomics is to tailor medications to an individual’s genetic makeup in order to improvethe benefit-to-risk ratio. Significant findings have been documented showing the effect of certain geneticvariations (e.g., in CYP2C9 and VKORC1) on the dose response to warfarin. Pharmacogenomic andgenetic information is crucial to improving the efficacy and safety of pharmacotherapy and for the optimalmanagement of thromboembolic disorders.

Key words: Pharmacogenomics, cardiovascular disease, cytochrome P450, warfarin, singlenucleotide polymorphisms.

1. Introduction

According to the American Heart Association, approximately 80million people (one out of three) have one or more forms of car-diovascular disease (CVD) (1). This number places a major bur-den on drug discovery efforts to improve the treatment of CVD.In today’s world of prescribed medications, doctors commonlyengage in the practice of trial and error. Consider, however, thepossibility of doctors being able to prescribe medications basedon a patient’s specific genetic makeup, knowing in advance howthe patient will respond (2, 3). This is the promise of pharma-cogenomics.

The term pharmacogenetics, first introduced by Vogel in1959 (4), is defined as the analysis of inherited factors that definean individual’s response to a drug. The term generally refers tothe identification and analysis of monogenetic variants that affect


277

278 Mousa

drug response. Pharmacogenomics, on the other hand, refers tothe identification and analysis of the entire library of genes thatimpact drug efficacy and safety. The most common genetic vari-ations are single nucleotide polymorphisms (SNPs), which occuron average at least once every 1,000 base pairs. SNPs account fornearly 3 million genetic variations and are distributed through-out the entire genome. Genetic variations that occur at a fre-quency of at least 1% in the human population are referred toas polymorphisms. Genetic polymorphisms are inherited and aremonogenic (i.e., involving one locus) and exhibit inter-ethnic dif-ferences in frequency. Rare mutations are those that occur at a fre-quency of less than 1% in the human population. Other examplesof genetic variations include insertion–deletion polymorphisms,tandem repeats, defective splicing, aberrant splice sites, and pre-mature stop codon polymorphisms.

There are many different factors that will contribute to howa patient responds to certain medications, such as age, gender,body weight, nutrition status, organ function status, presence ofinfections, and concomitant medications. We can now add geneticfactors to this list (5). In addition, there has been a recent shiftfrom looking at single genes (genetics) to focusing on the func-tion and interactions of the whole genome (6). A major focus ofcurrent research in the field of pharmacogenomics is on cardio-vascular medicine (7).

The goal of pharmacogenomics is personalized medicine, inwhich the type of drug and dosing regimen are tailored for eachindividual, as opposed to the one-drug-fits-all approach of mostcurrent medical practices. An important facet of the pharmacoge-nomic approach is being able to predict who will respond toa specific drug and who will experience adverse reactions (8).Even with the most advanced medications, not every patient hasa full response to every drug (9). A significant clinical problemwith current cardiovascular medications is adverse drug reactions,which are a major cause of hospitalization in the United Statestoday. Pharmacogenomics could help identify patients that willexperience adverse effects, thus avoiding potential toxicity andeven death (8). It is generally believed that individualized ther-apy based on pharmacogenomics will result in more effective,safer medications and more accurate dosing regimens, both ofwhich can lead to decreased health-care costs and improved cost-effectiveness by reducing hospitalizations due to adverse events,and decreasing the number of failed drug attempts in an effort tofind and establish an effective regimen (7).

With advances in genotyping technologies, we are now wellon our way to being able to understand how genetic variationscan impact treatment (8). The precise mechanism(s) underlyingthe variability in drug response among individuals is not clear;however, there is considerable evidence that genetic makeup is

Pharmacogenomics in Thrombosis 279

at least partially responsible (10). Most research in this area hasfocused on SNPs and DNA copy number variants (CNVs). Thegoal of such studies is to link SNPs or CNVs to the expression ofa target gene, which can be assessed using DNA microarrays, andthereby determine how an individual will respond to a given med-ication (8). One of the current challenges of pharmacogenomics isthat while we know that both genetic and non-genetic factors caninfluence an individual’s response to a particular drug, and thesefactors are beginning to be identified, it is not clear the degree towhich each factor effects variations in drug response (8).

One area of pharmacogenomics that has been incorporatedinto current clinical practice is genetic variation in the genes forthe cytochrome P450 (CYP) enzymes. CYP enzymes metabo-lize many classes of medications, including cardiovascular drugs.Variations in the genes that encode CYP enzymes can result inthe synthesis of overactive or inactive forms of the enzymes (seebelow). Patients who carry an inactive enzyme variant will beunable to metabolize and eliminate drugs properly, resulting indrug accumulation and possible serious toxicity (5). The inclu-sion of pharmacogenetic data on CYP polymorphisms has begunto be included in the packaging information that accompanies cer-tain drugs. The package insert for warfarin, for example, providesphysicians with genomic information related to drug dosing andhow individual responses to the drug may vary (11).

2. SNP Mapping

With the completion of the human genome, gene-basedapproaches using SNP markers have become an important toolin the identification of underlying causes, diagnosis, and treat-ment of disease. Through linkage disequilibrium (LD) analysis,or analysis of non-random association between SNPs in proxim-ity to each other, tens of thousands of anonymous SNPs can beidentified and mapped. These anonymous SNPs may fall withinnon-related genes, within susceptibility genes, or in non-codingsequences. Once identified and mapped, SNP markers can be usedto identify a region of the genome that harbors a putative “sus-ceptibility gene,” i.e., genetic variations that directly influencethe likelihood that an individual will develop disease. Positionalcloning and sequence analysis then identifies the gene and/or theSNP that underlies a specific condition or disease (12).

The concept of susceptibility genes has led to the identifica-tion of a number of putative gene variants associated with hema-tologic, haemostatic, and thrombotic disorders (Table 12.1).Notably, variations in the gene for thiopurine methyltransferase

280 Mousa

Table 12.1Genes involved in hematologic, haemostatic, and thrombotic disorders

Clone ID Name Description

810512 THBS1 Thrombospondin 1

205185 THBD Thrombomodulin143443 TBXAS1 Thromboxane A synthase 1

812276 SNCA Synuclein, alpha (non-A4 component of amyloid precursor)149910 SELL Selectin E (endothelial adhesion molecule 1)

753211 PTGER3 Prostaglandin E receptor 3 (subtype EP3)245242 CPB2 Carboxypeptidase B2 (plasma, carboxypeptidase U)

130541 PECAM1 Platelet/endothelial cell adhesion molecule (CD31 antigen)40643 PDGFRB Platelet-derived growth factor receptor, beta polypeptide

121218 PF4 Platelet factor 466982 PLGL Plasminogen like

810017 PLAUR Plasminogen activator, urokinase receptor813841 PLAT Plasminogen activator, tissue serine (or cystein) proteinase inhibitor,

Clade E (nexin, plasminogen activator inhibitor type 1), member 1

160723 LAMC1 Laminin, gamma 1 (formerly LAMB2)32609 LAMA4 Laminin, alpha 4

51447 FCGR3B Fc fragment of IgG, low-affinity IIIb, receptor for Z(CD16)41898 PTGDS Prostaglandin D2 synthase (21kD, brain)

810124 PAFAH1B3 Platelet activating factor acetylhydrolase, isoform 1b, gamma subunit(29kD)

810010 PDGFRL Platelet-derived growth factor receptor like

184038 SPTBN2 Spectrin, beta, non-erythrocytic 2179276 FASN Fatty acid synthase

776636 BHMT Betaine-homocystein methyltransferase770462 CPZ Carboxypeptidase Z

137836 PDCD10 Programmed cell death 10127928 HBP1 HMG-box containing protein 1

138991 COL6A3 Collagen, type VI, alpha 3212649 HRG Histidine-rich glycoprotein

155287 HSPA1A Heat shock 70 kD protein 1A810891 LAMA5 Laminin, alpha 5

811792 GSS Glutathione synthetase768246 G6PD Glucose-6-phosphate dehydrogenase

260325 ALB Albumin131839 FOLR1 Folate receptor 1 (adult)

139009 FN1 Fibronectin 1813757 FOLR2 Folate receptor 2 (fetal)

241788 FGB Fibrinogen, B beta polypeptide


Table 12.1 (continued)

Clone ID Name Description

49509 EPOR Erythropoietin receptor

26418 EDG1 Endothelial differentiation, sphingolipid G-protein-coupled receptor.1813254 F2R Coagulation factor II (thrombin) receptor

261519 TNFRSF5 Tumor necrosis factor receptor superfamily, member 5243816 CD36 CD36 antigen (collagen type 1 receptor, thrombospondin receptor)

714106 PLAU Plasminogen activator, urokinase758266 THBS4 Thrombospondin 4

712641 PRG4 Proteoglycan 4 (megakaryocyte stimulating factor, articular superficialzone protein)

726086 TFPI2 Tissue factor pathway inhibitor 2

67654 PDGFB Platelet-derived growth factor beta polypeptide (simian sarcoma viral(v-sis) oncogene homolog

85979 PLG Plasminogen

85678 F2 Coagulation factor II814378 SPINT2 Serine protease inhibitor, kunitz type 2

71101 PROCR Protein C receptor, endothelial (EPCR)785975 F13A1 Coagulation factor XIII, A1 polypeptide

310519 F10 Coagulation factor X840486 VWF Von Willebrand factor

191664 THBS2 Thrombospondin 2

(TPMT), a drug metabolizing enzyme, and 5-lipoxygenase(ALOX5) have been linked to adverse drug reactions and vari-ability in drug response, respectively (13, 14). An understandingof genetic variability is particularly important in the context ofsafety and efficacy of anticoagulant drugs.

2.1. SNPs in DrugMetabolizingEnzymes

Polymorphisms in drug metabolizing enzymes are the first recog-nized and most documented examples of genetic variations thathave consequences not only in drug response but also drug tox-icity. Drug metabolizing enzymes are divided into phase-I andphase-II metabolizing enzymes.

2.1.1. Phase-IMetabolizing Enzymes

Most phase-I enzymes are members of the CYP superfamily andlocalize to cells of the liver and gastrointestinal system. Approx-imately 40 different CYP enzymes are present in humans. Theyare classified according to family (e.g., 2), subfamily (e.g., D),and gene (e.g., 6) associated with the biotransformation (e.g.,CYP2D6). Functional genetic polymorphisms have been iden-tified for CYP2A6, CYP2C9, CYP2C19, and CYP2D6 (15),

282 Mousa

and more recently for CYP3A4 (16). A polymorphism in theregulatory region of the gene encoding CYP1A2 has been recog-nized, but its functional importance is unknown (17). CYP2CP,the primary pathway of metabolism of warfarin, metabolizes theS-isomer of the racemic warfarin mixture, which is 3 times morepotent than the R-isomer. The two most common CYP2C9 vari-ants, CYP2C9∗2 and CYP2C9∗3, have single amino acid substitu-tions at critical positions in the enzyme (18). CYP2C9∗3 homozy-gotes can exhibit as high as 90% reduction in S-warfarin clearanceas compared to wild-type carriers (19). CYP2C9 mutant alleleswere found to be overrepresented in 81% of patients requiringlow-dose warfarin therapy (<1.5 mg/d) (20). These patients typ-ically have difficulty in induction, require longer hospitalizationfor stabilization of the warfarin regimen and experience morebleeding complications. Patients that were homozygous for theCYP2C9∗3 allele had a profound response to warfarin, necessitat-ing a dose reduction to 0.5 mg/d.

It is now recognized that polymorphisms in the CYP genescan result in three possible drug phenotypes: poor, normal, andultrafast metabolizers. Poor metabolizers lack an active form ofthe expressed enzyme, normals have one copy of the active gene,and ultrafast metabolizers have duplicate copies of the active gene.Poor metabolism results in toxicity (the drug stays in the bodyfor a longer period of time), while ultrafast metabolism resultsin reduced drug efficacy. More than half of prescribed drugs aremetabolized by components of the CYP system. Of this half,CYP3A4 metabolism accounts for approximately 50%, CYP2D6accounts for 20%, and CYP2D9 and CYP2D19 account for 15%.In the future, drug developers will be able to identify and avoiddrugs whose metabolic pathways are significantly influenced bygenetic polymorphisms in the CYP system.

2.1.2. Phase-IIMetabolizing Enzymes

Examples of phase-II metabolizing enzymes that exhibit geneticpolymorphisms are N-acetyltransferase, TPMT, and glutathioneS-transferase. The relevance of genetic polymorphisms ofTPMT, dihydropyrimidine dehydrogenase (DPD) and UDP-glucuronosyl transferase (UGT) in cancer have been demon-strated (21–23). The TPMT gene has three mutant alleles,referred to as TPMT∗3A, TPMT∗2, and TPMT∗3C, with themost common being TPMT∗3A.

2.2. Drug TransporterGene Polymorphisms

Drug transport across the gastrointestinal lining, drug excretioninto the bile and urine, and drug distribution across the blood–brain barrier are mediated by specific membrane proteins. Geneticvariations in these proteins can cause disturbances in the distri-bution of a drug and alter drug concentrations, both of whichwill affect the therapeutic action and efficacy of the drug. TheP-glycoprotein complex is an energy-dependent transmembrane


efflux pump that is encoded by MDR-1 (multi-drug resistantgene 1). Fifteen MDR-1 polymorphisms have been identified todate. The most common, located in exon 26 of MDR-1 (24), hasbeen shown to influence P-glycoprotein expression in vitro (24).

3. Pharmacoge-nomics ofWarfarin Therapy

Some of the most promising pharmacogenomic data at presentfor CVD medications center on the anticoagulant warfarin. War-farin is the most commonly prescribed anticoagulant for the pre-vention and treatment of thromboembolic events (25). How-ever, warfarin is known to have a narrow therapeutic index, andinappropriately high doses can cause significant bleeding events.Thus, anticoagulation status in patients receiving warfarin mustbe monitored, particularly upon initiation of treatment, typicallythrough evaluation of prothrombin time and international nor-malized ratio (INR) (26). Warfarin is metabolized via the hep-atic CYP enzyme CYP2C9. Because of its narrow therapeuticindex, warfarin is highly sensitive to drugs that inhibit or enhanceCYP2C9 activity. In addition, genetic variations in CYP2C9 cancause some patients to metabolize warfarin more slowly, in whichcase the drug remains in the body for a longer period of time,putting the patient at an increased risk of serious bleeding (26).

There are two main variant alleles of CYP2C9 (Table 12.2),referred to as the ∗2 allele and the ∗3 allele (25). These alleles havebeen shown to cause reductions in CYP2C9 enzymatic activity byapproximately 30 and 80%, respectively, thus increasing the riskof warfarin-associated bleeding events (25). The mean warfarindose required in patients with the ∗3 or ∗2 allele is significantlylower than in those patients without these alleles. These find-ings are highly suggestive of a gene–dose relationship betweenCYP2C9 and warfarin (27). Variant alleles are associated withan increased time to reach stable dosing and an increased riskof above range INRs (INR > 3.2) as compared to wild-type car-riers (25, 26). Importantly, variant alleles are associated with anincreased risk of bleeding events, with the ∗3 allele having a higher

Table 12.2Polymorphisms associated with variable drug response to warfarin

Gene Polymorphisms Functional role Reference(s)

CYP2C9 ∗2 and ∗3 alleles Warfarin metabolism (20, 29, 30, 32, 34, 45)

VKORC1 −1639G>A and 1173C>T Activation of clotting factors (20, 29, 30, 32, 45)

284 Mousa

risk of bleeds (above range INRs) and a longer time to reachstable dosing as compared to the ∗2 allele. Ideally, the use ofgenotyping to guide dosing of warfarin will reduce the risk ofabove range INRs and bleeding events and reduce time to stabledosing (26).

The identification of genetic variants of the gene for the war-farin target, vitamin K epoxide reductase (encoded by VKORC1),has revealed another putative pharmacodynamic mechanism ofwarfarin resistance. Warfarin inhibits the activity of VKOR, result-ing in decreased levels of reduced vitamin K, which is required forthe function of several clotting factors (Fig. 12.1). Genetic vari-ations in VKORC1 that result in reduced activity of the enzyme,and thus decreased function of vitamin K-dependent clotting fac-tors, have been described (28). Several VKORC1 SNPs havebeen identified, but two in particular, at position –1639 ofthe VKORC1 promoter and +1173 in intron 1 of VKORC1,were found to correlate significantly with warfarin dose require-ments. For both of these positions, there was a significant risk ofincreased INRs with variant alleles (25, 27). Thus, in addition tosuch variables as age, sex, diet, and weight, genetic informationabout CYP2C9 and VKOR status may someday be considered indetermining warfarin dosing regimes for individual patients (27).

CYP2C9CYP2C9CYP1A1CYP1A1CYP1A2CYP1A2CYP3A4CYP3A4

RR--warfarin

warfarin

SS--warfarin

warfarin

Oxidized Vitamin KOxidized Vitamin K Reduced Vitamin KReduced Vitamin KOO22

HypofunctionalHypofunctionalF. II, VII, IX, XF. II, VII, IX, X

Protein C, S, ZProtein C, S, Z

Functional Functional F. II, VII, IX, XF. II, VII, IX, X

Proteins C, S, ZProteins C, S, Z

γ --glutamyl glutamyl carboxylasecarboxylase

Vitamin K Vitamin K ReductaseReductase

COCO22

WarfarinWarfarin

Calumenin

Fig. 12.1. Schematic representation of warfarin metabolism and the role of vitamin K reductase. The more potentS-warfarin is metabolized mainly via CYP2C9. Many clotting factors are dependent on the warfarin target, VKOR1. Poly-morphisms in CYP2C9 or VKOR1 may account in large part for inter-patient variability in warfarin response.


4. Pharmacoge-nomics in CVDand Beyond

This section summarizes important genetic mutations that havebeen shown to be associated with CVD and hematological dis-orders. Additional potentially relevant polymorphisms associatedwith CVD, general metabolism, and immune system function aresummarized in Table 12.3.

1. Coagulation factor II (prothrombin) G20210A. This muta-tion, present in 2% of the population, is located in the 3’untranslated region, near a putative polyadenylation site. Itis associated with increased levels of prothrombin, increasedrisk of deep vein thrombosis (DVT), recurrent miscarriagesand portal vein thrombosis in cirrhotic patients (29–32).

2. Coagulation factor V G1691A (Leiden R506Q). TheG1691A mutation, present in 8% of the population, isa specific G to A substitution at nucleotide +1691. The

Table 12.3Polymorphisms associated with CVD, metabolism and immune function

Gene Polymorphism(s) Association

Endothelial nitric oxidesynthase (eNOS)

E298D, G894T Increased risk for acute MI, coronaryatherosclerosis, and essential hypertension

Coagulation factor II(Prothrombin)

G20210A Elevated prothrombin levels with increasedrisk of DVT

Factor V (Leiden) G1691A Increased risk of venous thrombosis

Factor XIII V34L, G103T Lower incidence of CVDMethylenetetrahydrofolate

reductase (MTHFR)C677T Major risk factor for vascular disease

Apolipoprotein E E2–E3–E4 General metabolismButyrylcholinesterase atypic and K-variants General metabolism

Leukotriene C4 Synthase A-444c General metabolismIL1β C4336T (TaqI RFLP) Immune system and host immune response

IL4Rα Q576R Immune system and host immune responseIL6 G-174C Immune system and host immune response

TNFα G-238A, G-308A Immune system and host immune responseTNFβ A329G Immune system and host immune response

CD14 C-260T Immune system and host immune responseTLR4 Immune system and host immune response

CCR2-V641 (G190A) Immune system and host immune responseSDF-1 3’ G810A Immune system and host immune response

286 Mousa

resultant gene product is cleaved less efficiently (10%) byactivated protein C. The mutation is associated with DVT,recurrent miscarriage, portal vein thrombosis in cirrhoticpatients, early kidney transplant loss, and other forms ofvenous thromboembolism (29–31, 33), as well as a dramaticincrease in the incidence of thrombosis in women taking oralcontraceptives.

3. Coagulation factor IX propeptide missense mutations.Alanine-10 to threonine and alanine-10 to valine missensemutations in the factor IX propeptide result in abnormalsensitivity to oral anticoagulants through reduced affinity ofcarboxylase for the vitamin K-dependent coagulation fac-tor IX precursor, resulting in severe bleeding complica-tions (“coumarin hypersensitivity”). Patients with coumarinhypersensitivity experience severe bleeding, despite a thera-peutic INR. Increased activated partial thromboplastin time(aPTT) in these patients is due to reduced levels of factorIX. In the absence of coumarins, aPTT and factor IX lev-els are normal. Thus, analysis of nucleotide substitutions atposition +9311 (G>A threonine variant) or position +9312(C>T valine variant) are important in order to identify puta-tive hypersensitivity before initiation of coumarin therapy toavoid excessive bleeding complications (34–37).

4. Platelet glycoprotein Ia C807T. This gene polymorphism isassociated with nonfatal myocardial infarction in youngerpatients. Other platelet polymorphisms, such as those foundin P-selectin, α2 adrenergic receptor, and transforminggrowth factor-β (TNF-β) are also associated with increasedrisk of arterial disease or a prothrombotic state (38–41).

5. Platelet glycoprotein IIIa T393C (HPA-1 a/b, P1A1/A2).The glycoprotein IIIa P1A1/A2 alleles are associated witha leucine 33/proline 33 amino acid polymorphism and aredistinguishable by DNA typing and alloantibodies (42–44).HPA-1a (human platelet antigen-1a) was recently identifiedas an inherent risk factor for coronary thrombosis, prema-ture myocardial infarction, and coronary stent thrombosis.HPA-1a is also a determinant in the pathogenesis of post-transfusion purpura and neonatal alloimmune thrombocy-topenic purpura.

5. Summary andFuture Directions

Variability in drug response in different patients may be due togenetic differences that affect drug metabolism, drug distribu-tion, and drug target proteins (15). Variations in CYP enzymes


cause inter-individual differences in the plasma concentrations ofdrugs such as warfarin that have a narrow therapeutic index (20).In patients that experience adverse effects, medications may beeither avoided or administered at a reduced dose with carefulmonitoring, neither of which is ideal. Increased knowledge of theregulation of cellular functions and drug effects at the geneticlevel should lead to the development of new drugs and/or drugcombinations that are tailored to an individual’s needs. Throughthe discovery of new genetic targets, pharmacogenomics has thepotential to improve quality of life and reduce health-care costsby decreasing the number of treatment failures and adverse drugreactions.

Adverse drug reactions are the sixth leading cause of deathin the United States, with an annual expenditure of $75 billion.In addition to age, sex, and nutritional status, genetic factorsplay a significant role in individual’s response to a drug. Thereare a number of inherited sequence variants (alleles) of genesthat encode drug metabolizing enzymes and drug receptors thatmanifest as discrete drug metabolism phenotypes. Ideally, pre-treatment screening for these alleles, thereby enabling predictionsof a patient’s response to a medication, will help in selecting adrug and dosing regime that is safe and efficacious. The futuremay very well see physicians requesting a genotyping test insteadof a complete blood count, for example, in order to diagnosisand prescribe an individualized treatment. In this regard, recentadvances in molecular biological techniques for pin-pointing rel-evant mutations will be important.

Pharmacogenomics-based personalized medicine holds thepromise of reducing the number of adverse drug events and drugfailures, thereby vastly improving quality of life while reducinghealth-care costs. Pharmacogenomics also has the potential toprovide key insights for the diagnosis of drug resistance in throm-bosis and beyond.

References

1. Cardiovascular Disease Statistics. AmericanHeart Association, 2009 (Accessed January13, 2009, at http://www.americanheart.org/presenter.jhtml?identifier=4478).

2. Pharmacogenomics. Human Genome ProjectInformation, 2008. (Accessed January, 2009,at http://www.ornl.gov/sci/techresources/Human_Genome/medicine/pharma.shtml).

3. Aspinall, M., and Hamermesh, R. (2007)Realizing the promise of personalizedmedicine Harv Bus Rev 85, 108–17, 65.

4. Vogel, F. (1959) Moderne Probleme derHumangenetick. Ergebn Inn Med Klinder-heilk 12, 52–125.

5. Sadee, W., and Dai, Z. (2005) Pharmaco-genetics/genomics and personalized medi-cine Hum Mol Genet 14(Spec No. 2),R207–14.

6. Khoury, M. (2003) Genetics and genomics inpractice: the continuum from genetic diseaseto genetic information in health and diseaseGenet Med 5, 261–8.

7. Ginsburg, G., Donahue, M., and Newby, L.(2005) Prospects for personalized cardiovas-cular medicine: the impact of genomics J AmColl Cardiol 46, 1615–27.

8. Zhang, W., Huang, S., and Dolan, E.(2008) Integrating epigenomics into

288 Mousa

pharmacogenomic studies. PharmacogenomPersonal Med 1, 7–14.

9. Momary, K. (2007) Cardiovascular pharma-cogenomics J Pharm Prac 2007 20, 265–76.

10. Cresci, S. (2008) From SNPs to functionalstudies in cardiovascular pharmacogenomicsMethods Mol Biol 448, 379–93.

11. Warfarin drug label revisions. 2007 (AccessedJanuary 13, 2009, at http://www.ama-assn.org/ama1/pub/upload/mm/464/warfarin_label_revis.pdf).

12. Collins, F.S. (1992) Positional cloning: let’snot call it reverse anymore Nat Genet 1,3–6.

13. Drazen, J.M., Yandava, C.N., Dube, L.,Szczerback, N., Hippensteel, R., Pillari, A.,Israel, E., Schork, N., Silverman, E.S., Katz,D.A., and Drajesk, J. (1999) Pharmacoge-netic association between ALOX5 promotergenotype and the response to anti-asthmatreatment Nat Genet 22, 168–70.

14. Krynetski, E.Y., and Evans, W.E. (1999)Pharmacogenetics as a molecular basis forindividualized drug therapy: the thiopurineS-methyltransferase paradigm Pharm Res 16,342–9.

15. Evans, W.E., and Relling, M.V. (1999)Pharmacogenomics: translating functionalgenomics into rational therapeutics Science286, 487–91.

16. Sata, F., Sapone, A., Elizondo, G., Stocker,P., Miller, V.P., Zheng, W., Raunio, H.,Crespi, C.L., and Gonzalez, F.J. (2000)CYP3A4 allelic variants with amino acid sub-stitutions in exons 7 and 12: evidence foran allelic variant with altered catalytic activ-ity Clin Pharmacol Ther 67, 48–56.

17. Sachse, C., Brockmoller, J., Bauer, S., andRoots, I. (1999) Functional significance ofa C–>A polymorphism in intron 1 of thecytochrome P450 CYP1A2 gene tested withcaffeine Br J Clin Pharmacol 47, 445–9.

18. Stubbins, M.J., Harries, L.W., Smith, G.,Tarbit, M.H., and Wolf, C.R. (1996)Genetic analysis of the human cytochromeP450 CYP2C9 locus Pharmacogenetics 6,429–39.

19. Takahashi, H., Kashima, T., Nomoto,S., Iwade, K., Tainaka, H., Shimizu, T.,Nomizo, Y., Muramoto, N., Kimura, S.,and Echizen, H. (1998) Comparisonsbetween in-vitro and in-vivo metabolism of(S)-warfarin: catalytic activities of cDNA-expressed CYP2C9, its Leu359 variant andtheir mixture versus unbound clearance inpatients with the corresponding CYP2C9genotypes Pharmacogenetics 8, 365–73.

20. Aithal, G.P., Day, C.P., Kesteven, P.J., andDaly, A.K. (1999) Association of polymor-

phisms in the cytochrome P450 CYP2C9with warfarin dose requirement and risk ofbleeding complications Lancet 353, 717–9.

21. Ando, Y., Saka, H., Asai, G., Sugiura, S.,Shimokata, K., and Kamataki, T. (1998)UGT1A1 genotypes and glucuronidation ofSN-38, the active metabolite of irinotecanAnn Oncol 9, 845–7.

22. Lu, Z., Zhang, R., Carpenter, J.T., andDiasio, R.B. (1998) Decreased dihydropy-rimidine dehydrogenase activity in a popu-lation of patients with breast cancer: impli-cation for 5-fluorouracil-based chemotherapyClin Cancer Res 4, 325–9.

23. Relling, M.V., Hancock, M.L., Rivera,G.K., Sandlund, J.T., Ribeiro, R.C., Krynet-ski, E.Y., Pui, C.H., and Evans, W.E.(1999) Mercaptopurine therapy intoleranceand heterozygosity at the thiopurine S-methyltransferase gene locus J Natl CancerInst 91, 2001–8.

24. Hoffmeyer, S., Burk, O., von Richter, O.,Arnold, H.P., Brockmoller, J., Johne, A.,Cascorbi, I., Gerloff, T., Roots, I., Eichel-baum, M., and Brinkmann, U. (2000)Functional polymorphisms of the humanmultidrug-resistance gene: multiple sequencevariations and correlation of one allele withP-glycoprotein expression and activity in vivoProc Natl Acad Sci USA 97, 3473–8.

25. Anderson, J., Horne, B., Stevens, S., Grove,A., Barton, S., Nicholas, Z., Kahn, S.,May, H., Samuelson, K., Muhlestein, J.,and Carlquist, J. (2007) Randomized trialof genotype-guided versus standard warfarindosing in patients initiating oral anticoagula-tion Circulation 116, 2563–70.

26. Higashi, M., Veenstra, D., Kondo, L.,Wittkowsky, A., Srinouanprachanh, S.,Farin, F., and Rettie, A. (2002) Associ-ation between CYP2C9 genetic variantsand anticoagulation-related outcomes dur-ing warfarin therapy JAMA 287, 1690–8.

27. Sconce, E., Khan, T., Wynne, H., Avery,P., Monkhouse, L., King, B., Wood, P.,Kesteven, P., Daly, A.K., and Kamali,F. (2005) The impact of CYP2C9 andVKORC1 genetic polymorphism and patientcharacteristics upon warfarin dose require-ments: proposal for a new dosing regimenBlood 106, 2329–33.

28. Rieder, M., Reiner, A., Gage, B., Nickerson,D., Eby, C., McLeod, H., Blough, D.,Thummel, K., Veenstra, D., and Rettie, A.(2005) Effect of VKORC1 haplotypes ontranscriptional regulation and warfarin doseN Engl J Med 352, 2285–93.

29. Amitrano, L., Brancaccio, V., Guardascione,M.A., Margaglione, M., Iannaccone, L.,


D’Andrea, G., Marmo, R., Ames, P.R., andBalzano, A. (2000) Inherited coagulationdisorders in cirrhotic patients with portal veinthrombosis Hepatology 31, 345–8.

30. Foka, Z.J., Lambropoulos, A.F., Saravelos,H., Karas, G.B., Karavida, A., Agorastos,T., Zournatzi, V., Makris, P.E., Bontis, J.,and Kotsis, A. (2000) Factor V leiden andprothrombin G20210A mutations, but notmethylenetetrahydrofolate reductase C677T,are associated with recurrent miscarriagesHum Reprod 15, 458–62.

31. Manucci, P.M. (2000) The molecularbasis of inherited thrombophilia Vox Sang78(Suppl 2), 39–45.

32. Soria, J.M., Almasy, L., Souto, J.C., Tirado,I., Borell, M., Mateo, J., Slifer, S., Stone,W., Blangero, J., and Fontcuberta, J. (2000)Linkage analysis demonstrates that the pro-thrombin G20210A mutation jointly influ-ences plasma prothrombin levels and risk ofthrombosis Blood 95, 2780–5.

33. Ekberg, H., Svensson, P.J., Simanaitis, M.,and Dahlback, B. (2000) Factor V R506Qmutation (activated protein C resistance) isan additional risk factor for early renal graftloss associated with acute vascular rejectionTransplantation 69, 1577–81.

34. Chu, K., Wu, S.M., Stanley, T., Stafford,D.W., and High, K.A. (1996) A mutation inthe propeptide of Factor IX leads to warfarinsensitivity by a novel mechanism J Clin Invest98, 1619–25.

35. Harbrecht, U., Oldenburg, J., Klein, P.,Weber, D., Rockstroh, J., and Hanfland,P. (1998) Increased sensitivity of factor IXto phenprocoumon as a cause of bleedingin a patient with antiphospholipid antibodyassociated thrombosis J Intern Med 243,73–7.

36. Oldenburg, J., Quenzel, E.M., Harbrecht,U., Fregin, A., Kress, W., Muller, C.R.,Hertfelder, H.J., Schwaab, R., Brackmann,H.H., and Hanfland, P. (1997) Missensemutations at ALA-10 in the factor IX propep-tide: an insignificant variant in normal life buta decisive cause of bleeding during oral anti-coagulant therapy Br J Haematol 98, 240–4.

37. Stanley, T.B., Humphries, J., High, K.A.,and Stafford, D.W. (1999) Amino acidsresponsible for reduced affinities of vitamin

K-dependent propeptides for the carboxylaseBiochemistry 38, 15681–7.

38. Bray, P.F. (2000) Platelet glycoprotein poly-morphisms as risk factors for thrombosisCurr Opin Hematol 7, 284–9.

39. Kunicki, T.J., Kritzik, M., Annis, D.S., andNugent, D.J. (1997) Hereditary variation inplatelet integrin alpha 2 beta 1 density isassociated with two silent polymorphisms inthe alpha 2 gene coding sequence Blood 89,1939–43.

40. Reiner, A.P., Kumar, P.N., Schwartz,S.M., Longstreth, W.T., Jr., Pearce, R.M.,Rosendaal, F.R., Psaty, B.M., and Siscovick,D.S. (2000) Genetic variants of platelet gly-coprotein receptors and risk of stroke inyoung women Stroke 31, 1628–33.

41. Santoso, S., Kunicki, T.J., Kroll, H., Haber-bosch, W., and Gardemann, A. (1999) Asso-ciation of the platelet glycoprotein Ia C807Tgene polymorphism with nonfatal myocar-dial infarction in younger patients Blood 93,2449–53.

42. Newman, P.J., Derbes, R.S., and Aster,R.H. (1989) The human platelet alloanti-gens, PlA1 and PlA2, are associated witha leucine33/proline33 amino acid polymor-phism in membrane glycoprotein IIIa, andare distinguishable by DNA typing J ClinInvest 83, 1778–81.

43. Williamson, L.M., Hackett, G., Rennie, J.,Palmer, C.R., Maciver, C., Hadfield, R.,Hughes, D., Jobson, S., and Ouwehand,W.H. (1998) The natural history of feto-maternal alloimmunization to the platelet-specific antigen HPA-1a (PlA1, Zwa) asdetermined by antenatal screening Blood 92,2280–7.

44. Zotz, R.B., Winkelmann, B.R., Nauck, M.,Giers, G., Maruhn-Debowski, B., Marz, W.,and Scharf, R.E. (1998) Polymorphism ofplatelet membrane glycoprotein IIIa: humanplatelet antigen 1b (HPA-1b/PlA2) is aninherited risk factor for premature myocardialinfarction in coronary artery disease ThrombHaemost 79, 731–5.

45. Chasman, D., and Adams, R.M. (2001)Predicting the functional consequences ofnon-synonymous single nucleotide polymor-phisms: structure-based assessment of aminoacid variation J Mol Biol 307, 683–706.

Chapter 13

Diagnosis and Management of Sickle Cell Disorders

Shaker A. Mousa and Mohamad H. Qari

Abstract

Sickle cell disease (SCD) is a wide-spread inherited hemolytic anemia that is due to a point mutationleading to a valine/glutamic acid substitution in the β-globin chain, causing a spectrum of clinical man-ifestations in addition to hemolysis and anemia. Acute painful crisis is a common sequela that can causesignificant morbidity and negatively impact the patient’s quality of life. Remarkable improvements in ourunderstanding of the pathogenesis of this clinical syndrome and the role of cell adhesion, inflammation,and coagulation in acute painful crisis have led to changes in the management of pain. Due to the endemicnature of SCD in various parts of the Middle East, a group of physicians and scientists from the UnitedStates and Middle East recently met to draw up a set of suggested guidelines for the management ofacute painful crisis that are reflective of local and international experience. This chapter brings togethera detailed etiology, pathophysiology, and clinical presentation of SCD, including the differential diag-noses of pain associated with the disease, with evidence-based recommendations for pain managementand the potential impact of low-molecular weight heparin (LMWH), from the perspective of physiciansand scientists with long-term experience in the management of a large number of SCD patients.

Key words: Painful crisis, vaso-occlusive crisis, sickle cell, guidelines, low-molecular weight heparin,prophylaxis, diagnosis, management.

1. Introduction

The inherited disorders of hemoglobin are the most commongene disorders, and it is estimated that 7% of the world’s popula-tion are carriers. Approximately 300,000 children worldwide areborn with documented sickle cell disease (SCD) every year. Sick-ling disorders are found frequently in the Afro-Caribbean pop-ulations and sporadically throughout the Mediterranean region,India, and the Middle East (1).


291

292 Mousa and Qari

SCD is due to a genetic mutation that results in an aminoacid substitution of valine for glutamic acid at position 6 ofthe adult β-globin chain. In the homozygous form, this sub-stitution results in polymerization of hemoglobin upon deoxy-genation, leading to deformed dense red blood cells in sicklecell anemia (SCA) patients. The predominant pathophysiologi-cal feature of homozygous SCA is vaso-occlusion, which leads toacute and chronic complications such as painful crises, increasedrisk of infection, acute chest syndrome, stroke, and severe painepisodes with dramatic hemolysis, which progress to involve mul-tiple organs, including the central nervous system, cardiovascularsystem, lung, liver, bone, skin, and kidneys (2).

The epidemiology of SCA in the Arab region and the highprevalence of the mutation are well documented. In Saudi Arabia,the overall prevalence of the sickle cell trait is 4.20%, with widevariation from region to region. In Oman, the prevalence of thesickle cell trait is 6%, and in Bahrain, 11.2% of the population aresickle cell gene carriers (3–5).

Extensive studies conducted over several years in differentprovinces of Saudi Arabia have revealed a wide distribution pat-tern of the hemoglobin S (HbS) gene in different provinces. Theoverall prevalence of HbS carrier status in Saudi Arabia varies from0–1% in the northern and central regions to approximately 7% inthe western, 12% in the southern, and nearly 25% in some partsof the eastern region (6, 7).

The gene frequency of HbS in the populations of the Gulfregion and other middle eastern states is one of the highest in theworld. This chapter will present a consensus opinion on the man-agement of painful vaso-occlusive crisis based on the recommen-dations of a panel of physicians and scientists from these regions.The objective is to provide guidelines for the local practicinghematologist and non-hematologist who might not be exposedto large numbers of SCA patients, with the goal of standardizingpreventive immediate care in the emergency room, upon admis-sion, and upon follow-up. The serious morbidity and hamperedquality of life inflicted by painful crisis creates an urgent need forwell-informed guidelines for the management and treatment ofpatients suffering from SCD at the local level.

2. Pathogenesis

The clinical hallmark of SCA is the painful acute “crisis,” whichdespite therapeutic advances, continues to be a treatment chal-lenge. Such crises occur with variable frequency and duration, andthey commonly require hospitalization (8).

Diagnosis and Management of Sickle Cell Disorders 293

The pathobiology of SCD is characterized by episodicvascular occlusion, with multiple inciting events beyond actualHbS polymerization and mechanical obstruction induced by sick-led red blood cells (RBCs). An important mechanism of paininduction is thought to involve bone marrow vasculature infarc-tion, leading to the release of inflammatory mediators that inturn stimulate afferent nerve fibers and cause pain. Vaso-occlusionalso involves adherence of circulating blood elements, such asleukocytes, to endothelial cells, hyper-coagulability, endothe-lial dysfunction, altered nitric oxide metabolism, and ischemiareperfusion injury (9–11).

2.1. Increased RedCell Adhesion

Abnormal interactions between erythrocytes and the endotheliumare a primary initiating factor in the development of microvascularocclusions in SCD. This view is supported by reports of a signif-icant correlation between clinical severity of SCD and extent ofRBC adhesiveness (9, 12).

2.2. IncreasedLeukocyte Adhesion

Neutrophils are likely to be an important factor in causingmicrovascular sickle cell trapping and consequent vaso-occlusion.Sickle cells appear to be more adherent to neutrophils than nor-mal RBCs. Sickle cells also induce neutrophil oxidative activity,which might be important in neutrophil-induced tissue damageduring vaso-occlusive episodes (13).

2.3.Hyper-Coagulability

Patients with SCD exhibit high plasma levels of markers ofthrombin generation, depletion of natural anticoagulant proteins,reduced activity of the fibrinolytic system, and increased tissuefactor expression, even in the non-crisis steady state. Plateletsand other cellular elements are also chronically activated, whichmight predispose the patient to thromboembolic manifestations(14, 15). Because of the increased generation of thrombin andfibrin and the increased tissue factor pro-coagulant activity inpatients, SCD is often described as a hyper-coagulable state(16, 17). Recently, it was shown that compared to healthy age-and race-matched controls, patients with SCD had higher levelsof plasma markers of coagulation activation (D-dimers, F1+2, andthrombin-anti-thrombin III complex) (18).

2.4. ReperfusionInjury and NitricOxide

A central aspect of sickle cell vasculopathy is the impairmentof endothelial regulation of vasomotor tone, thrombosis, andinflammation (9). Intermittent vascular occlusion in SCA patientsleads to reperfusion injury, which is associated with granulocyteaccumulation and the enhanced production of reactive oxygenspecies. The recruitment of nitric oxide (NO) to counteractthe resultant oxidative stress reactions results in a reduction inNO bioavailability and contributes to vascular dysfunction inSCD (10).

294 Mousa and Qari

2.5. Roleof InflammatoryMediators

There is an emerging consensus that a pro-inflammatory statecontributes to the vaso-occlusive complications associated withSCD. Tissue damage due to vaso-occlusion results in the releaseof numerous inflammatory mediators that initiate the trans-mission of painful stimuli and the perception of pain. Plasmacytokines and related factors represent a burgeoning area ofinquiry related to the pathogenesis in SCD. Cytokines derivedfrom platelets, white blood cells, and endothelial cells havebeen implicated in the development of several sequelae of thedisease (19, 20). In as much as sickle cell adhesion to theendothelium plays a role in vaso-occlusion, the existence of suchdiverse mechanisms of adhesion presents an enormous challengein terms of identifying physiologically relevant therapeutic tar-gets. Interestingly, a recent study has suggested that targetinga specific adhesion pathway may be sufficient to reduce vaso-occlusion (21).

In summary, the vascular pathophysiology of SCD is dueto the combination of many factors, including red and whiteblood cell adhesion to the endothelium, increased coagulation,and homeostatic perturbation. The vascular endothelium is cen-tral to disease pathogenesis because it presents adhesion moleculesfor the attachment of blood cells, balances the pro-coagulant andanticoagulant properties of the vessel wall, and regulates vascularhomeostasis by synthesizing vaso-constricting and vaso-dilatingsubstances. Intermittent vascular occlusion in SCD leads to reper-fusion injury, which is associated with granulocyte accumula-tion and enhanced production of reactive oxygen species. The

Table 13.1Rational for tinzaparin in SCDa

Tinzaparin:

• Exerts favorable pharmacodynamic effects as compared to other LMWHs on a variety of cel-lular factors, including endothelial TFPI, vWF, TNF-α, NO modulation, and P/L-selectin

• Inhibits matrix degrading enzymes more effectively than other LMWHs

• Reduces oxidative stress by inhibiting the generation of reactive oxygen species• Induces long-lasting increases in plasma TFPI and NO

• Has favorable anticoagulant efficacy and safety profiles• Has a once daily pharmacokinetic profile

• Does not require any special dose adjustment or monitoring in special patient populations,including severe renal failure patients

asee Refs. (8, 45–55).LMWH, low-molecular weight heparin; TFPI, tissue factor pathway inhibitor; vWF, vonWillebrand factor; TNF,tumor necrosis factor; NO, nitric oxide; SCD, sickle cell disease.


recruitment of NO to counteract the resultant oxidative stressreactions results in reduced NO bioavailability and contributes tovascular dysfunction in SCD.

Table 13.1 illustrates the impact of the low molecular weightheparin (LMWH) tinzaparin on many of these complex processes.

3. ClinicalPresentationof the PainfulCrisis

3.1. FactorsPrecipitating PainfulCrisis

Pain can be precipitated by hypoxia, infection, fever, acidosis,dehydration, pregnancy, menstruation, obstructive sleep apnea,and exposure to cold or abrupt weather changes. Coinheritanceof the methylenetetrahydrofolate reductase (MTHFR) C677Tmutation also predisposes SCA patients to pain (22). Patientsalso cite anxiety, depression, alcohol consumption, and physicalexhaustion. Pain can be precipitated by co-morbidities such assarcoidosis, diabetes mellitus, cholecystitis, and herpes; in manyinstances, no precipitating event is identified (23–25).

3.2. Effects of Painon Quality of Lifeand Survival

SCD is chronic and lifelong. Individuals are most often well, buttheir lives are punctuated by periodic painful attacks, and theyare at increased risk of other medical complications and prema-ture death (26). Several studies have examined the relationshipbetween pain and quality of life, and current emphasis in treat-ment is on managing distress, pain relief, and psychological sup-port (27).

The average life expectancy of males and females with SCDis 42 and 48, respectively (26). Individuals who experience >6episodes per year have a reduced survival rate compared to thosewho experience less frequent events. However, with good healthcare, many individuals with SCA maintain reasonably good healthmost of the time and live productive lives. In fact, in the past 30years, the life expectancy of individuals with SCA has increased(23).

3.3. Characteristicsof Pain

The acute painful crisis is characterized by a sudden onset ofpain that may start in any part of the body. The pain is variablefrom mild to severe with excruciating deep pain that is felt inthe bones and soft tissues. Acute painful crisis is distinct fromother varieties of pain that can arise due to complications of sicklecell anemia, such as acute chest syndrome, priapism, splenic andhepatic sequestration, hand foot syndrome, arthritis, and abdom-inal pain caused by calcular cholecystitis. Before embarking on a

296 Mousa and Qari

diagnosis of acute painful crisis, it is important that these othercauses of pain be ruled out (23).

4. Managementof PainfulEpisodes

Figures 13.1 and 13.2 present summary guidelines for managingpainful crisis in adults and children, respectively. These guidelines

Adjuvant therapy:1. Hydroxyurea2. Tinzaparin (Inno hep®) (175 IU/kg daily x–27

days)3. Ibuprofen, Paracetemol (PO or IV),

Diclofenac, Ketorlac4. Laxative: Lactose (10 ml 1.0 bid)5. Antiemetic: Prochloroperazine (5–10 mg PO

tid) or Metochlopramide 6. Anxiolytic: Haloperidol (1–3 mg PO or I/M bid

in case of anxiety or agitation)7. Fentanyl (25 mcg, patch will last for 3 days)

Pain is severe and not responding to oral analgesia(pain assessment tools might include visual analog scale, verbal scale etc.)

Rapid clinical assessment, clinical laboratory chemistry, blood film, blood gases

Insert IV line/encourage oral fluid intake

Morphine (0.1 mg/kg IV or SC every 20 minutes)

Maintenance:Morphine (0.05–0.1 mg/kg/2–4hrs SC/IV or PO; alternatively, use PCA) 1. Diamorphin (0.01 mg/kg IV or SC); or

Hydromorphine 2. Rule out other causes of severe pain3. Involve pain management team and social

team 4. Consider transfer to ICU5. Exchange transfuse the patient and consider

caudal analgesic if the patient has lower body pain

(PCA should be the guidance for opioid administration to minimize adverse effects)

Pain controlled Pain persists

Monitor pain, vital signs and O2 saturation every 30 min;when pain is controlled, repeat monitoring every 2 hours

If respiratory rate <10/min or O2saturation <90% or patient sedated, give Naloxone (0.1 mg repeated every 2 min as necessary) and stop sedatives

Pain improved based on standardized pain assessment tools; discharge patient and arrange home care plan with oral analgesics

Fig. 13.1. Management of adults with painful crisis. IV, intravenous; SC, subcutaneous; PO, per oral; PCA, patient-controlled analgesia; bid, twice daily; tid, thrice daily; IM, intramuscular.


Mild to moderate pain: Paracetamol (15 mg/Kg/dose)

+ Codeine (1 mg/kg/dose PO q4 hours)Ibuprofen (5–10 mg/kg/dose PO q6–8 hours; max 2400

mg/day)

Pain Persists

Moderate to severe pain: Morphine (0.1–0.15 mg/kg/dose, repeat hourly)

Adjuvant therapy:1. Hydroxyurae2. Tinzaparin (200–240 IU/kg SC once daily)3. Ibuprofen, Paracetemol (PO or IV), Diclofenac, Ketorlac4. Laxative: Lactulose (10 ml 1.0 bid)5. Antiemetic: Prochloroperazine (5–10 mg PO tid)6. Anxiolytic: Haloperidol (1–3 mg PO or I/M bid)

Maintenance: additional 0.05 mg/kg Morphine q1–2 hours until improvement.

Change to oral analgesia, Paracetamol or Ibuprofen

Discharge and arrange home care plan

Improvement

Rapid clinical assessmentTests (Table 2)

Insert IV line/encourage oral fluid intake

±

Fig. 13.2. Management of painful crisis in children (less than 11 years of age). IV, intravenous; SC, subcutaneous; PO,per oral; bid, twice daily; tid, thrice daily; IM, intramuscular.

298 Mousa and Qari

have been drawn up based on the recommendations andexperience of a panel of international basic and clinical scientistsand local practitioners from several major hospitals in Saudi Arabiaand the Gulf region, as well as a number of excellent reviewsdetailing the clinical management of painful crisis in SCA patients(11, 24, 28–43).

4.1. Managementat Presentation

Patients should be seen immediately by a physician and athorough history and examination should be carried out. Routinelaboratory testing might not be necessary in patients with uncom-plicated vaso-occlusive crisis. If a patient has symptoms that aresevere enough to warrant hospitalization, laboratory tests shouldinclude a complete blood count, reticulocyte count, and urinal-ysis. If fever is present, a chest radiograph should be obtained,and urine, sputum, and blood should be cultured to rule outinfection. Fever is common in patients with uncomplicated vaso-occlusive crisis and does not necessarily indicate the presence ofan underlying infection (33). If the patient experiences severeabdominal pain, recurrent vomiting, respiratory symptoms, neu-rological signs of paresis or paralysis, acute joint swelling, pri-apism, or an abrupt fall in hemoglobin, the treating physicianshould be alerted, and the etiology should be identified andtreated (23).

Comprehensive approaches to ambulatory and inpatient painmanagement have been addressed in a number of publications(20, 29, 33). These reports are a particularly good resource forestablishing clinical pathways for hospital-based pain managementin patients with SCD. Recommendations include the timely useof opioid medications for moderate to severe pain and avoidanceof meperidine (Demerol) (11, 30–33, 37).

4.2. Home-BasedVersus HospitalTreatment

The best approach to managing pain is aggressive analgesictherapy with frequent reassessment of effectiveness using a stan-dardized pain assessment tool. The treatment of milder episodesof pain can be achieved at home with oral fluids, oral anal-gesics such as paracetamol, non-steroidal anti-inflammatory drugs(NSAIDs) (ibuprofen in particular for children) (37), and com-fort measures, such as heating pads. For the patient who has per-sistent pain, codeine or tramadol is recommended. When home-based management fails to adequately alleviate pain, it is essen-tial that patients undergo rapid triage, physical assessment, andaggressive, appropriately monitored analgesia (31).

4.3. Hospitalization For severe pain, the patient will require hospitalization. Hydrationand parenteral opioids such as morphine are indicated and usuallyadministered by scheduled around-the-clock dosing or patient-controlled analgesia (PCA) (29, 31, 33). During admissionand episodes of severe pain, life-threatening complications maydevelop rapidly and often are heralded by relatively sudden clinical


Table 13.2Routine clinical investigation at presentation

• Complete blood count and differential

• Reticulocyte count• Blood chemistry profile

• Urine analysis and cultures• Chest X-ray

• Pulse oxymetry• In pregnant patients the following tests are also required:

◦ Hemoglobin electrophoresis◦ Screening for viral hepatitis

◦ Vaginal swabs and endocervical smears◦ Blood grouping and antibody screening

◦ Iron studies (TIBC, ferritin)◦ Folate and vitamin 12 levels

◦ Thyroid function test

changes, such as increased oxygen requirement, altered mentalstatus, or decreased hemoglobin levels or platelet counts (39, 41).Guidelines for routine clinical measures to be taken at presenta-tion are summarized in Table 13.2.

Opioids should not be withheld because of the unfoundedfear of addiction. A recent retrospective evaluation of pain assess-ment and treatment for acute vaso-occlusive episodes in childrenwith SCD concluded that despite opiate dosing within recom-mended guidelines, mean pain scores remained in the moderateto severe range for several days following hospitalization for vaso-occlusive episodes (35).

The fact that pain is severe and protracted and involvescomplex underlying etiologies, including impaired cell–cell inter-actions, increased production of pro-inflammatory mediators,hemostatic imbalance, and hyper-coagulability, paves the way forthe development of alternative approaches to therapy, includingthe use of anti-adhesion, anti-inflammatory, and anticoagulationagents. Recently, the LMWH tinzaparin, given as a supplementto opioid treatment, was found to significantly shorten the pro-tracted course of pain in a randomized, controlled, double-blindstudy (8). Thus, tinzaparin is justified for use in the treatmentof acute painful crisis in SCA. The clinical effects of tinzaparinare most likely due to its favorable pharmacodynamic effectson a variety of cellular factors, including endothelial tissue fac-tor pathway inhibitor (TFPI), tumor necrosis factor (TNF)-α,NO activity, von Willebrand factor (vWF), matrix degrad-ing enzymes, and P-selectin (44–53). Long-lasting increases in

300 Mousa and Qari

plasma concentrations of TFPI and NO levels induced by tin-zaparin in particular would contribute favorable anticoagulanteffects. Given once daily, tinzaparin does not require any spe-cial monitoring (8, 54, 55). However, further trials are requiredbefore making a general recommendation.

4.4. PotentialProphylaxicand TherapeuticImpact ofAnti-thromboticAgents on SCD

Aspirin, standard heparin, and warfarin have been used forthe treatment of acute painful crisis without conclusive results(44, 46). Tinzaparin has been show to have anti-thrombotic, anti-inflammatory, and vascular protective effects in obese and healthysubjects and is justified in the treatment of acute painful crisis inSCA based on a randomized, double-blind clinical trial (8, 48,51–53) (see Table 13.1).

4.5. AncillaryMeasures

Ancillary measures to reduce the incidence of complications dur-ing acute painful crisis include the use of frequent incentivespirometry while the patient is awake to encourage deeper inspi-ratory effort, and avoidance of fluid overload by limiting overallintake to 1.0–1.5 times the maintenance need (56).

Oxygen supplementation is not needed unless hypoxemia ispresent. Close monitoring of oxygen saturation and respiratorystatus, with particular attention to excessive sedation, is neces-sary (57). Other adjuvants include antihistamines, anticoagulants,antidepressants, benzodiazepines, and anticonvulsants. This is aheterogeneous group of compounds that potentiate the analgesicand ameliorate the side effects of opioids, and exert mild anal-gesic effects. The role of selective serotonin reuptake inhibitorsin SCA is not clear at present. Adjuvants must be used with care,and patients should be monitored carefully when receiving them.Adjuvant therapy can also have adverse effects, some of whichprecipitate or worsen the manifestations of SCA.

Other considerations include maintaining adequate (but notexcessive) hydration, the provision of heating pads, massages,warm baths, and other comfort measures, monitoring of oxy-genation and cardiopulmonary status, and close observation ofother potential complications, particularly acute chest syndrome(56–63).

4.6. DischargeCriteria

The patient is considered ready for discharge when he/she cantolerate oral fluids and medications, pain is controlled by POmedication, and concurrent problems are resolved. Long hospitalstays are often complicated by hospital-acquired infections.

4.7. Rehabilitationand PsychologicalConsiderations in theManagement of SCD

Recurrent pain has an immeasurably negative impact on dailyactivities, school and work performance, social interactions andrelationships, mood, quality of life, and recreation activities. Thepsychosocial aspects of chronic pain are complex and often are


Table 13.3Recommended physical and psychosocial services for SCA patients

• Establishment of a dedicated unit for the care of SCD

• Establishment of a dedicated multi-disciplinary team of specialized doctors, nurses, physical therapist,psychologist/psychiatrist, social workers, pain management specialist

• Good spectrum of pain medication drugs

• Physiotherapy• TENS therapy

• Behavioral/cognitive therapy• Participation in national societies/support groups

• Acupuncture/acupressure may be beneficial• Distraction and entertainment, including videos and other media

• Improved clinical outcomes in the management of SCD crisis• Medical and health education of medical staff and patient on:

◦ Forms of and preventing exposure to hypoxia (i.e., altitude, smoking)◦ Habits and the use of charcoal in heating within closed areas

◦ Avoidance of excessive physical stress and competitive exercise◦ Avoidance of sudden changes in temperature

◦ Avoidance of exposure to infection◦ Avoidance of dehydration

SCA, sickle cell anemia; TENS, transcutaneous electrical nerve stimulation

not appropriately addressed. A patient with chronic and recur-rent pain must undergo a thorough psychosocial assessment tohelp define stressors and co-morbidities (i.e., depression), copingstrategies, and support systems (64).

Many patients with SCD suffer from physical and psycho-logical stressors/complications due to the disease or diseasetherapy. These complications can include divorce, loss of fam-ily ties and friendship, unemployment due to interruption ofwork, dependence on social welfare support, and stereotyp-ing of patient problems with drug-seeking behavior. Gener-ally, there is no consistent plan for disease management andfollow-up by physicians who are not experienced in the treat-ment of SCD. This lack of a consistent plan, in additionto the disabling complications of SCD such as chronic painsyndrome, avascular necrosis, arthropathies, and other skeletalmanifestations, can result in depression and low self-esteem.Table 13.3 presents an example of a well-coordinated structuredprogram based on the physical and psychological aspects of SCD(64–67).

302 Mousa and Qari

5. Pregnancyand Sickle CellAnemia

Pregnancy increases the complications of SCA for the mother,fetus, and neonate, but there have been significant improvementsin outcomes (68–70). The corner stone of management duringpregnancy involves the hematologist and the obstetrician, as themother needs regular monthly follow-ups and close fetal moni-toring. All pregnant SCA patients should receive folic acid fromthe beginning of pregnancy. Immunization history should be cur-rent, particularly hepatitis B, pneumococcal, and influenza vacci-nations. Asymptomatic bacteruria must be treated as well as anyother active infection. Patient must be instructed to adequatelyhydrate and avoid physical and psychological stress.

5.1. Managementof Sickle CellPatients DuringPregnancy

Forty-eight percent of women with SCD experience a crisisduring pregnancy (69, 71). Painful sickle cell crisis requires admis-sion and treatment with intravenous hydration and pain control.Analgesia should be adjusted based on the patient’s response,with rapid relief being the goal. NSAIDs are not the therapyof choice in pregnant patients who have a mild crisis becauseof the concern for premature closure of the ductus arteriosusat advanced gestational age. Rather, opiates, preferably mor-phine, are recommended. Patients should be on a scheduleddosage with additional boluses if needed. Oxygen therapy shouldbe given if oxygen saturation is less than the patient’s knownsteady state. Adjuvant therapy such as stool softeners, antiprurit-ics, and anxiolytics/sedatives should be considered (72). Bloodtransfusion may be indicated if signs or symptoms of anemiaare present (tachycardia, tachypnea, dyspnea, fatigue, decreasinghemoglobin, a low reticulocyte count of <100×109/L). Bloodshould be leukopoor and antigen matched. In extreme cases,exchange transfusions may be required (73).

5.2. LMWHin Pregnancyand Sickle CellAnemia

LMWHs are regarded as attractive alternatives to unfractionatedheparin (UFH) as anticoagulants during pregnancy due to theirclinical advantages and their association with a lower incidenceof bleeding, osteoporosis, and heparin-induced thrombocytope-nia (HIT). Many published reviews confirm that LMWHs area safe alternative to UFH as anticoagulants during pregnancy(74–79).

Tinzaparin sodium, the active ingredient of innohep R©, hasbeen used extensively in pregnant women at risk of developingor with a history of venous thromboembolism (VTE) and hasproven to be a safe and effective pharmacologic agent. A recentmulticentre, prospective, dose-finding clinical study report pub-lished in the American Journal of Obstetrics and Gynaecology


documented the safety and efficacy of innohep R© treatment ofhigh-risk pregnancy cases of various aetiologies by also measur-ing anti-Xa levels throughout the gestation period and beyond(80). A treatment dose of 175 IU/kg/day and a prophylacticdose of 75 IU/kg/day reliably achieved target activities that weresufficient and safe, and there were stable gestational tinzaparinrequirements (no dose adjustments). In addition, there were noapparent clinical effects of treatment on bone density.

The use of tinzaparin during crises in SCA patients has beendocumented (8); however, the issue of its use during crises inpregnant SCA patients has not been thoroughly investigated.Recent guidelines acknowledge that SCA pregnant patients areat risk of developing VTE and, accordingly, a prophylactic doseof LMWH can be given throughout the pregnancy and post-partum (81).

5.3. ManagementDuring Labor

Generally, delivery can be accomplished vaginally, with cesareansection reserved for obstetric indications. Spontaneous labor ispreferable because sickle cell crisis has been reported to be asso-ciated with induction of labor. There is a concern that the use ofprostaglandins may lead to cell sickling, and therefore, should beused with caution (73).

5.4. BloodTransfusion

Prophylactic blood transfusions have been used to treat pregnantwomen who have sickle disorders. Currently, their use is contro-versial. A comparative trial of prophylactic versus on demand-based transfusion showed no difference in prenatal outcomebetween the two groups. The group that was transfused hada lower incidence of painful crisis; however, other medical andobstetric complications occurred with equal frequency (73).

6. Conclusions

SCD, with its burden of excruciating pain associated with sicklecell crisis, is the most common globin gene disorder and a majorhealth problem. The prediction in very early childhood of latersevere disease could justify the early use of disease-modifying pro-cedures or interventions, such as hydroxyurea treatment, chronictransfusions, or stem cell transplantation. The hazards of thesetreatments vary greatly, particularly as compared to preventiveand supportive management alone. Accurate and reliable earlypredictors would provide the opportunity to better balance therisks of these interventions with risks of the disease itself andmight reduce the frequency of hospitalization and blood trans-fusion, the incidence of pain, and the occurrence of acute chest

304 Mousa and Qari

syndrome and pulmonary hypertension in patients with SCD.A randomized, double-blind clinical study of the LMWH tinza-parin demonstrated significant and rapid pain resolution, under-scoring the urgent need for additional therapeutic strategiesdesigned to counteract the endothelial dysfunction and theinflammatory, hypercoagulation, and oxidative abnormalities ofSCD. Currently, there is an unmet medical need for better man-agement of SCD patients. Not covered by this review, but ofspecial consideration in the management of SCD patients, arerenal failure, addiction to opiates, and peri-operative painful crisismanagement.

References

1. Jeremiah, Z.A. (2006) Abnormal haemo-globin variants, ABO and Rh blood groupsamong student of African descent in PortHarcourt, Nigeria Afr Health Sci 6, 177–81.

2. Hiran, S. (2005) Multiorgan dysfunctionsyndrome in sickle cell disease J Assoc PhysicIndia 53, 19–22.

3. Alhamdan, N.A., Almazrou, Y.Y., Alswaidi,F.M., and Choudhry, A.J. (2007) Premaritalscreening for thalassemia and sickle cell dis-ease in Saudi Arabia Genet Med 9, 372–7.

4. Al-Riyami, A., and Ebrahim, G.J. (2003)Genetic blood disorders survey in the Sul-tanate of Oman J Trop Pediatr 49(Suppl 1),i1–20.

5. Mohammed, A.M., Al-Hilli, F., Nadkarni,K.V., Bhagwat, G.P., and Bapat, J.P.(1992) Hemoglobinopathies and glucose-6-phosphate dehydrogenase deficiency in hos-pital births in Bahrain Ann Saudi Med 12,536–9.

6. el-Hazmi, M.A., and Warsy, A.S. (1999)Appraisal of sickle-cell and thalassaemia genesin Saudi Arabia East Mediterr Health J 5,1147–53.

7. El-Hazmi, M.A., and Warsy, A.S. (1994) Thefrequency of glucose-6-phosphate dehydro-genase phenotypes and sickle cell genes inAl-Qatif oasis Ann Saudi Med 14, 491–4.

8. Qari, M.H., Aljaouni, S.K., Alardawi, M.S.,Fatani, H., Alsayes, F.M., Zografos, P.,Alsaigh, M., Alalfi, A., Alamin, M., Gadi, A.,and Mousa, S.A. (2007) Reduction of painfulvaso-occlusive crisis of sickle cell anaemia bytinzaparin in a double-blind randomized trialThromb Haemost 98, 392–6.

9. Aslan, M., and Freeman, B.A. (2007) Redox-dependent impairment of vascular functionin sickle cell disease Free Radic Biol Med 43,1469–83.

10. Hebbel, R.P., Osarogiagbon, R., andKaul, D. (2004) The endothelial biology of

sickle cell disease: inflammation and a chronicvasculopathy Microcirculation 11, 129–51.

11. Mehta, S.R., Afenyi-Annan, A., Byrns, P.J.,and Lottenberg, R. (2006) Opportunities toimprove outcomes in sickle cell disease AmFam Physician 74, 303–10.

12. Hebbel, R.P., Boogaerts, M.A., Eaton, J.W.,and Steinberg, M.H. (1980) Erythrocyteadherence to endothelium in sickle-cell ane-mia. A possible determinant of disease sever-ity N Engl J Med 302, 992–5.

13. Grigg, A.P. (2001) Granulocyte colony-stimulating factor-induced sickle cellcrisis and multiorgan dysfunction in apatient with compound heterozygoussickle cell/beta+thalassemia Blood 97,3998–9.

14. Ataga, K.I., and Key, N.S. (2007) Hyper-coagulability in sickle cell disease: newapproaches to an old problem Hematol AmSoc Hematol Educ Program 2007, 91–6.

15. Inwald, D.P., Kirkham, F.J., Peters, M.J.,Lane, R., Wade, A., Evans, J.P., and Klein,N.J. (2000) Platelet and leucocyte activationin childhood sickle cell disease: associationwith nocturnal hypoxaemia Br J Haematol111, 474–81.

16. Tomer, A., Harker, L.A., Kasey, S., andEckman, J.R. (2001) Thrombogenesis insickle cell disease J Lab Clin Med 137,398–407.

17. Key, N.S., Slungaard, A., Dandelet, L.,Nelson, S.C., Moertel, C., Styles, L.A.,Kuypers, F.A., and Bach, R.R. (1998) Wholeblood tissue factor procoagulant activity iselevated in patients with sickle cell diseaseBlood 91, 4216–23.

18. Ataga, K.I., Moore, C.G., Hillery, C.A.,Jones, S., Whinna, H.C., Strayhorn, D.,Sohier, C., Hinderliter, A., Parise, L.V.,and Orringer, E.P. (2008) Coagulationactivation and inflammation in sickle cell


disease-associated pulmonary hypertensionHaematologica 93, 20–6.

19. Etienne-Julan, M., Belloy, M.S., Decastel,M., Dougaparsad, S., Ravion, S., and Hardy-Dessources, M.D. (2004) Childhood sicklecell crises: clinical severity, inflammatorymarkers and the role of interleukin-8 Haema-tologica 89, 863–4.

20. Ballas, S.K. (2007) Current issues in sicklecell pain and its management Hematol AmSoc Hematol Educ Prog 2007, 97–105.

21. Frenette, P.S., and Atweh, G.F. (2007)Sickle cell disease: old discoveries, new con-cepts, and future promise J Clin Invest 117,850–8.

22. Al-Absi, I.K., Al-Subaie, A.M., Ameen, G.,Mahdi, N., Mohammad, A.M., Fawaz,N.A., and Almawi, W.Y. (2006) Associa-tion of the methylenetetrahydrofolate reduc-tase A1298C but not the C677T singlenucleotide polymorphism with sickle cell dis-ease in Bahrain Hemoglobin 30, 449–53.

23. Mid-Atlantic Regional Human Genetics Net-work (MARHGN). Sickle cell disease in chil-dren and adolescents: diagnosis, guidelinesfor comperhensive care, and protocols formanagement of acute and chronic compli-cations. In: Mid-Atlantic Sickle Cell DiseaseConsortium (MASCC) Practice GuidelinesWorkgroup;2002.

24. Shapiro, B.S. (1989) The management ofpain in sickle cell disease Pediatr Clin NorthAm 36, 1029–45.

25. Platt, O.S., Thorington, B.D., Brambilla,D.J., Milner, P.F., Rosse, W.F., Vichinsky, E.,and Kinney, T.R. (1991) Pain in sickle celldisease. Rates and risk factors N Engl J Med325, 11–6.

26. Platt, O.S., Brambilla, D.J., Rosse, W.F.,Milner, P.F., Castro, O., Steinberg, M.H.,and Klug, P.P. (1994) Mortality in sickle celldisease. Life expectancy and risk factors forearly death N Engl J Med 330, 1639–44.

27. Barakat, L.P., Patterson, C.A., Daniel, L.C.,and Dampier, C. (2008) Quality of lifeamong adolescents with sickle cell disease:mediation of pain by internalizing symptomsand parenting stress Health Qual Life Out-comes 6, 60.

28. Hassell, K.L., Eckman, J.R., and Lane, P.A.(1994) Acute multiorgan failure syndrome:a potentially catastrophic complication ofsevere sickle cell pain episodes Am J Med 96,155–62.

29. Robieux, I.C., Kellner, J.D., Coppes,M.J., Shaw, D., Brown, E., Good, C.,O’Brodovich, H., Manson, D., Olivieri,N.F., Zipursky, A., et al. (1992) Anal-gesia in children with sickle cell crisis:

comparison of intermittent opioids vs.continuous intravenous infusion of mor-phine and placebo-controlled study of oxy-gen inhalation Pediatr Hematol Oncol 9,317–26.

30. Yale, S.H., Nagib, N., and Guthrie, T. (2000)Approach to the vaso-occlusive crisis in adultswith sickle cell disease Am Fam Physic 61,1349–56, 63–4.

31. Benjamin, L., Dampier, C., Jacox, A.,Odesina, V., Pheonix, D., Shapiro, B.,Strafford, M., and Treadwell, M. Guidelinefor the Management of Acute and ChronicPain in Sickle-Cell Disease. American PainSociety;1999.

32. American Academy of Pediatrics, S.o.H.O.C.o.G.A.A.o. (2002) Health supervision forchildren with sickle cell disease Pediatrics109, 526–35.

33. Preboth, M. (2000) Management of painin sickle cell disease Am Fam Physician 61,1544, 9–50.

34. Reid, C., Charache, S., and Lubin, B.e. Man-agement of Therapy of Sickle Cell Disease,3rd edition. Bethesda, MD: National Insti-tutes of Health;1995.

35. Jacobson, S.J., Kopecky, E.A., Joshi, P., andBabul, N. (1997) Randomised trial of oralmorphine for painful episodes of sickle-celldisease in children Lancet 350, 1358–61.

36. Zempsky, W.T., Loiselle, K.A., McKay, K.,Blake, G.L., Hagstrom, J.N., Schechter,N.L., and Kain, Z.N. (2008) Retrospectiveevaluation of pain assessment and treatmentfor acute vasoocclusive episodes in childrenwith sickle cell disease Pediatr Blood Cancer51, 265–8.

37. Panel, S.C.D.G. Sickle Cell Disease: Screen-ing, Diagnosis, Management and Counsel-ing in Newborns and Infants. Rockville, MD:AHCPR;1993.

38. Field, J.J., Knight-Perry, J.E., and Debaun,M.R. (2009) Acute pain in children andadults with sickle cell disease: managementin the absence of evidence-based guidelinesCurrent Opin Hematol 16, 173–8.

39. Morrissey, L.K., Shea, J.O., Kalish, L.A.,Weiner, D.L., Branowicki, P., and Heeney,M.M. (2009) Clinical practice guidelineimproves the treatment of sickle cell diseasevasoocclusive pain Pediatr Blood Cancer 52,369–72.

40. Niscola, P., Sorrentino, F., Scaramucci, L.,de Fabritiis, P., and Cianciulli, P. (2009) Painsyndromes in sickle cell disease: an updatePain Med(Malden, Mass 10), 470–80.

41. Pack-Mabien, A., and Haynes, J., Jr. (2009)A primary care provider’s guide to preven-tive and acute care management of adults and

306 Mousa and Qari

children with sickle cell disease J Am AcadNurse Practition 21, 250–7.

42. Richard, R.E. (2009) The management ofsickle cell pain Curr Pain Headache Rep 13,295–7.

43. Meremikwu, M.M. (2009) Sickle cell diseaseClin Evid (Online) 2009, 2402.

44. Chaplin, H., Jr., Monroe, M.C., Malecek,A.C., Morgan, L.K., Michael, J., andMurphy, W.A. (1989) Preliminary trial ofminidose heparin prophylaxis for painfulsickle cell crises East Afr Med J 66, 574–84.

45. Embury, S.H., Matsui, N.M., Ramanujam,S., Mayadas, T.N., Noguchi, C.T., Diwan,B.A., Mohandas, N., and Cheung, A.T.(2004) The contribution of endothelial cellP-selectin to the microvascular flow ofmouse sickle erythrocytes in vivo Blood 104,3378–85.

46. Kirkham, F.J., Lerner, N.B., Noetzel, M.,DeBaun, M.R., Datta, A.K., Rees, D.C., andAdams, R.J. (2006) Trials in sickle cell diseasePediatr Neurol 34, 450–8.

47. Matsui, N.M., Varki, A., and Embury, S.H.(2002) Heparin inhibits the flow adhesion ofsickle red blood cells to P-selectin Blood 100,3790–6.

48. Mousa, S.A. (2005) Elevation of plasma vonWillebrand factor and tumor necrosis factor-ain obese subjects and their reduction by thelow molecular weight heparin tinzaparin IntAngiol 24, 278–81.

49. Mousa, S.A. (2005) Effect of low molec-ular weight heparin and different heparinmolecular weight fractions on the activityof the matrix-degrading enzyme aggrecanase:structure-function relationship J Cell Biochem95, 95–8.

50. Mousa, S.A. (2006) Inhibitory effect ofC-reactive protein on the release of tissue fac-tor pathway inhibitor from human endothe-lial cells: reversal by low molecular weightheparin Int Angiol 25, 10–3.

51. Mousa, S.A., Bozarth, J., and Barrett, J.S.(2003) Pharmacodynamic properties of thelow molecular weight heparin, tinzaparin:effect of molecular weight distribution onplasma tissue factor pathway inhibitor inhealthy human subjects J Clin Pharmacol 43,727–34.

52. Mousa, S.A., and Johansen, K. (2005) Phar-macodynamic effects of low molecular weightheparin in obese subjects following subcuta-neous administration of 75 IU/kg on plasmatissue factor pathway inhibitor and nitricoxide Int Angiol 24, 40–2.

53. Stevenson, J.L., Choi, S.H., and Varki, A.(2005) Differential metastasis inhibitionby clinically relevant levels of heparins–

correlation with selectin inhibition, notantithrombotic activity Clin Cancer Res 11,7003–11.

54. Mousa, S.A. (2002) The low molecularweight heparin, tinzaparin, in thrombosis andbeyond Cardiovasc Drug Rev 20, 199–216.

55. Puskas, A., Balogh, Z., Hadadi, L., Imre, M.,Orban, E., Kosa, K., Brassai, Z., andMousa, S.A. (2007) Spontaneous recanaliza-tion in deep venous thrombosis: a prospec-tive duplex ultrasound study Int Angiol 26,53–63.

56. Bellet, P.S., Kalinyak, K.A., Shukla, R.,Gelfand, M.J., and Rucknagel, D.L. (1995)Incentive spirometry to prevent acute pul-monary complications in sickle cell diseasesN Engl J Med 333, 699–703.

57. Vichinsky, E.P., Neumayr, L.D., Earles, A.N.,Williams, R., Lennette, E.T., Dean, D.,Nickerson, B., Orringer, E., McKie, V.,Bellevue, R., Daeschner, C., and Manci, E.A.(2000) Causes and outcomes of the acutechest syndrome in sickle cell disease. NationalAcute Chest Syndrome Study Group N EnglJ Med 342, 1855–65.

58. Greenberg, J., Ohene-Frempong, K.,Halus, J., Way, C., and Schwartz, E. (1983)Trial of low doses of aspirin as prophylaxis insickle cell disease J Pediatr 102, 781–4.

59. Osamo, N.O., Photiades, D.P., and Famodu,A.A. (1981) Therapeutic effect of aspirinin sickle cell anaemia Acta Haematol 66,102–7.

60. Quinn, C.T., and Buchanan, G.R. (1999)The acute chest syndrome of sickle cell dis-ease J Pediatr 135, 416–22.

61. Salvaggio, J.E., Arnold, C.A., and Banov,C.H. (1963) Long-term anti-coagulation insickle-cell disease. A clinical study N Engl JMed 269, 182–6.

62. Wolters, H.J., ten Cate, H., Thomas, L.L.,Brandjes, D.P., van der Ende, A., van derHeiden, Y., and Statius van Eps, L.W. (1995)Low-intensity oral anticoagulation in sickle-cell disease reverses the prethrombotic state:promises for treatment? Br J Haematol 90,715–7.

63. Zago, M.A., Costa, F.F., Ismael, S.J., Tone,L.G., and Bottura, C. (1984) Treatment ofsickle cell diseases with aspirin Acta Haema-tol 72, 61–4.

64. Ballas, S.K. (1990) Treatment of pain inadults with sickle cell disease Am J Hematol34, 49–54.

65. Anie, K.A., and Green, J. (2002) Psycholog-ical therapies for sickle cell disease and painCochrane Database Syst Rev CD001916.

66. Bodhise, P.B., Dejoie, M., Brandon, Z.,Simpkins, S., and Ballas, S.K. (2004)


Non-pharmacologic management of sicklecell pain Hematology 9, 235–7.

67. Thomas, V.J., Gruen, R., and Shu, S. (2001)Cognitive-behavioural therapy for the man-agement of sickle cell disease pain: identifica-tion and assessment of costs Ethn Health 6,59–67.

68. Serjeant, G.R. (1997) Sickle-cell diseaseLancet 350, 725–30.

69. Sun, P.M., Wilburn, W., Raynor, B.D., andJamieson, D. (2001) Sickle cell disease inpregnancy: twenty years of experience atGrady Memorial Hospital, Atlanta, GeorgiaAm J Obstet Gynecol 184, 1127–30.

70. Villers, M.S., Jamison, M.G., De Castro,L.M., and James, A.H. (2008) Morbidityassociated with sickle cell disease in preg-nancy Am J Obstet Gynecol 199, 125 e1–5.

71. Smith, J.A., Espeland, M., Bellevue, R.,Bonds, D., Brown, A.K., and Koshy, M.(1996) Pregnancy in sickle cell disease: expe-rience of the cooperative study of sickle celldisease Obstet Gynecol 87, 199–204.

72. Rees, D.C., Olujohungbe, A.D., Parker,N.E., Stephens, A.D., Telfer, P., andWright, J. (2003) Guidelines for the manage-ment of the acute painful crisis in sickle celldisease Br J Haematol 120, 744–52.

73. Koshy, M., Burd, L., Wallace, D.,Moawad, A., and Baron, J. (1988) Pro-phylactic red-cell transfusions in pregnantpatients with sickle cell disease. A random-ized cooperative study N Engl J Med 319,1447–52.

74. Faron, G., Corbisier, C., Tecco, L., andVokaer, A. (2001) First sickle cell crisis

triggered by induction of labor in aprimigravida Eur J Obstet Gynecol Reprod Biol94, 304–6.

75. Sanson, B.J., Lensing, A.W., Prins, M.H.,Ginsberg, J.S., Barkagan, Z.S., Lavenne-Pardonge, E., Brenner, B., Dulitzky, M.,Nielsen, J.D., Boda, Z., Turi, S., MacGillavry, M.R., Hamulyak, K., Theunissen,I.M., Hunt, B.J., and Buller, H.R. (1999)Safety of low-molecular-weight heparin inpregnancy: a systematic review ThrombHaemost 81, 668–72.

76. Bates, S.M., Greer, I.A., Hirsh, J., andGinsberg, J.S. (2004) Use of antithromboticagents during pregnancy: the seventh ACCPconference on antithrombotic and throm-bolytic therapy Chest 126, 627–44S.

77. Ginsberg, J.S., Greer, I., and Hirsh, J. (2001)Use of antithrombotic agents during preg-nancy Chest 119, 122–31S.

78. Greer, I.A. (2005) Venous thromboem-bolism and anticoagulant therapy in preg-nancy Gend Med 2(Suppl A), S10–7.

79. Villa-Forte Gomes, M.P. (2009) Venousthromboembolism in pregnancy Curr Treat-ment Opt Cardiovasc Med 11, 104–13.

80. Smith, M.P., Norris, L.A., Steer, P.J.,Savidge, G.F., and Bonnar, J. (2004) Tinza-parin sodium for thrombosis treatment andprevention during pregnancy Am J ObstetGynecol 190, 495–501.

81. Gynaecologists, R.C.o.O.a. Thromboem-bolic Disease in Pregnancy and the Puer-perium: Acute Management. GuidelineNo. 28 Update. London: RCOG Press;2007.

SUBJECT INDEX

A

Abciximabclinical utility. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .268heparin-induced thrombocytopenia (HIT). . . .207–208metastasis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254percutaneous coronary intervention . . . . . . . . . . . . . . . 771

Activated factor VIIa, see Tissue factor (TF)/activatedfactor VIIa

Activated partial thromboplastin time (aPTT). . . . . . . .2–3,5, 51, 66, 69, 81–83, 88, 142–143, 147,165–166, 168–170, 174, 236, 242, 248, 286

Activated protein C (APC) . . . . . . . . . . . . . . . . . 44, 163, 286Acute myocardial infarction (AMI)

animal models of . . . . . . . . . . . . . . . . . . . . . . . . . . 33–34, 91low molecular weight heparin (LMWH) . . . . . . . . . . . 33reperfusion therapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

Acute pulmonary embolism . . . . . . . . . . . 115, 158, 182, 188Tinzaparin . . . . . . . . . . . . . . . . . . . . . . . 299–300, 302–303

Adverse drug reactions . . . . . . . . . . . . . . . . . . . . 278, 281, 287Pharmacogenomics . . . . . . . . . . . . . . . . . . . . 278–279, 287

Alanine aminotransferase . . . . . . . . . . . . . . . . . . . . . . .173, 184Elevated . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173, 184

Alpha 2 integrin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85Alpha 4 integrin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265–267

Alpha 4 beta 1 (VLA-1) . . . . . . . . . . . . . . . 264, 266–267Alpha 4 beta . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7, 265–266

Alpha 5 beta 1 integrin . . . . . . . . . . . . . . . . . . . . . . . . . 267, 271Angiogenesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267bacterial infection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267

Alpha 5 beta 3 integrin . . . . . . . . . . . . . . . . . . . . 262, 268–270Alpha 5 beta 5 mediated adhesion assay . . . . . . . . . . 270–271Alpha 5 beta l–receptor–biotinylated fibronectin–binding

assay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271AMI, see Acute myocardial infarction (AMI)3-aminopropyltriethoxysilane treated glass slides . . . . . . . 23

platelet immobilization on . . . . . . . . . . . . . . . . . . . . . . . . 23preparation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

Angiogenesisalpha 5 beta 1 integrin . . . . . . . . . . . . . . . . . . . . . . . . . . 267coagulation . . . . . . . . .111, 114–115, 119–120, 122–125,

127, 161–162, 232, 243, 248low molecular weight heparin (LMWH) . . . . . 110–111tissue factor (TF) . . . . . . . . . . . . . . . . . . . . . . . . . . 110–111,

115, 122, 243tissue factor (TF)/activated factor VIIa. . . . . . . . . . . .110tumor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120, 123,

125, 243Animal thrombosis models

clinical correlates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91clinical relevance . . . . . . . . . . . . . . . . . . . . . . . . . . 30, 33, 88selection of . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32–33, 89

Animal venous thrombosis modelschemically-induced thrombosis models . . . . . . . . . . . . 54laser-induced thrombosis models . . . . . . . . . . . . . . . . . . 58stasis-thrombosis model . . . . . . . . . . . . . . . . . . . . . . . 31, 58vena-caval ligation model . . . . . . . . . . . . . . . . . . . . . . . . . 47vessel-wall damage models . . . . . . . . . . . . . . . . . 31–32, 64

AnticoagulantsDuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50, 89heparin-induced thrombocytopenia (HIT). . . . . . . .136,

140, 142–144, 146–147, 158, 171, 236, 302latest developments . . . . . . . . . . . . . . . . . . . . . . 47, 51, 123,

142, 146, 235–236tumor factors predicting sensitivity to . . . . . . . . . . . . . 111

Antifactor Xa agents . . . . . . . . . . . . . . . . . . . . . . . . . . 111–113,163–164, 175, 194, 248, 303

Antigen assays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136–138Antiheparin-PF4 antibody . . . . . . . . . . . . 133–138, 140–141

enzyme-linked immunosorbent assays(ELISA) . . . . . . . . . . . . . . . . . . . . . . . 137–138, 148

Anti-integrins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 262potential drug discovery target . . . . . . . . . . . . . . . . . . . 262

Antiplatelet efficacy assays . . . . . . . . . . . . . . . . . . . . . . 269–271Antiplatelets

ADP receptor antagonists . . . . . . . . . . . . . . . . . . . 204, 216bleeding risks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222, 226natural . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231

Anti-platelet therapyacute coronary syndrome (ACS) . . . . . . . . . . . . . 187, 204combination . . . . . . . . . . . . . . . . . 125, 204, 207–215, 226

aspirin and clopidogrel . . . . . . . . . . . . . . 204, 208–215GPIIb/IIIa antagonists . . . . . . . . . . . . . . . . . . . . . . 207

Anti-selectins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 262–264Antithrombin III (ATIII) . . . . . . . . . . . . . 111, 115–116, 235Antithrombotic agents

Novel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33, 51, 81, 88bleeding tendency . . . . . . . . . . . . . . . . . . . . . . . . . 88–89thrombosis animal models . . . . . . . . . . . . . . . . . 88–89discovery . . . . . . . . . . . . . . . . . . . . . . . . . 33, 89, 92, 204

Antithrombotics . . . . . . . . 1–26, 29–94, 142, 144, 147–148,158–159, 162, 164, 166, 168, 170, 204, 216,229–237

Antitissue factor . . . . . . . . . . . . . . . . . 110, 161–162, 251, 255Anti-Xa agents . . . . 111–113, 163–164, 175, 194, 248, 303APC, see Activated protein C (APC)Argatroban (Novastan)

acute venous thromboembolism. . . . . . . . . . . . . . .88, 144antithrombotic efficacy . . . . . . . . . . . . . . . . . . . . . . . . 88, 92co-therapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147heparin-induced thrombocytopenia (HIT) . . . 143–144,

158, 172platelet aggregation . . . . . . . . . . . . . . . . . . . . . . . . . . 12, 149Thrombosis . . . . . . . . . . . . . . . . . . . . . . . . .88, 91, 143–144

S.A. Mousa (ed.), Anticoagulants, Antiplatelets, and Thrombolytics, Methods in Molecular Biology 663,DOI 10.1007/978-1-60761-803-4, c© Springer Science+Business Media, LLC 2003, 2010

309

310ANTICOAGULANTS, ANTIPLATELETS, AND THROMBOLYTICS

Subject Index

Arterial thrombosis . . . . . . . . . . . . .44, 46–48, 53, 56–57, 82,161–163, 165, 169, 181, 203, 222, 230, 236

copper coil-induced canine . . . . . . . . . . . . . . . . . . . . . . . 62Arterial and venous thrombosis model

(Harbauer) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45–48Arteriovenous shunt thrombosis model . . . . . . . . . . . . . . . . 64Aspirin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8, 12, 36, 38,

40, 44, 47, 51, 59, 77, 79, 88, 127, 147, 167,184, 187, 194, 196, 203–215, 222–223, 230,235, 254, 268–269, 300

heparin-induced thrombocytopenia (HIT) . . . . . . . . 147Atherosclerotic plaques . . . . . . . . . . . . . . . . . . . . . . . . . . 48, 213

adhesion molecules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269ATIII, see Antithrombin III

B

Bacterial infection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267alpha 4 beta 1 integrin (VLA-1) . . . . . . . . . . . . . . . . . 266alpha 5 beta 1 integrin . . . . . . . . . . . . . . . . . . . . . . . . . . 267

Bernard-Soulier syndrome (BSS) . . . . . . . . . . . . . . . . . . . . . 24Beta 1 integrins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266–267

alpha 4 beta 1 (VLA-1) . . . . . . . . . . . . . . . . . . . . . . . . . 266alpha 5 beta . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1, 266bacterial infection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267gastrointestinal disease . . . . . . . . . . . . . . . . . . . . . . . . . . 267

Beta 2 integrins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267Beta 3 integrins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268–269

Alpha IIb beta 3 (GPIIb/IIIa) . . . . . . . . . . . . . . . 268–269antagonists . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268–269clinical utility . . . . . . . . . . . . . . . . . . . . . . . . . . .268–269

Alpha vs. beta . . . . . . . . . . . . . . . . . . . . . . . . . . . 3, 268–269Bivalirudin (Hirulog) . . . . . . . . . . . . . . . . . . . . . . 142, 145, 172Bleeding . . . . . . . . . . . . . . 30–32, 43, 46, 51, 57, 76–89, 120,

143–145, 147, 158, 162–170, 173–174,183–184, 186–198, 205–207, 209–215, 222,224–226, 230, 234, 236, 242, 250–251, 257,268–269, 282–284, 286, 302

animal models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76–79difficulties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88, 282

BSs . . . . . . . . . . . . . . . . . . . . . . . see Bernard-Soulier syndrome

C

CAM, see Cell adhesion molecules (CAMs)Cancer

Angiogenesis . . . . . . . . . . . 111, 120, 122–127, 162, 232,243, 248, 262, 269

anti-platelet inhibitors . . . . . . . . . . . . . . . . . 244, 254, 256coagulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122–124

tissue factor (TF) . . . . . . . . . . . . . . . . . . . . . . . 123–124complications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125, 236

deep vein thrombosis (DVT) . . . . . . . . . . . . . . . . . 136experimental models . . . . . . . . . . . . . . 111, 122, 162, 243heparin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110, 118low molecular weight heparin (LMWH) . . . . . . . . . . 110metastasis, lung . . . . . . . . . . . . . . . . . . . . . . . 122, 162, 244non-anticoagulant low molecular weight

heparin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247–248platelet-cancer cell adhesion . . . . . . . . . . . . . . . . . . . . . 120thrombosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110, 118–125tissue factor . . . . . . . . . . . . . 110, 121–122, 158, 242–243tissue factor pathway inhibitor

(TFPI) . . . . . . . . . . . . . . . . . . . . . . . . 110–111, 122venous thromboembolism (VTE) . . . . . . . . . . . . 120–122

Canine arterial thrombosis . . . . . . . . . . . . . . . . . . . . . . . . . . . 62copper coil-induced . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62

Canine coronary artery thrombosis model . . . . . . . . . . 55–57Canine injury-induced (electrolytic) arterial

thrombosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91CAPRIE trial . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215Cardiac surgery . . . . . . . . . . . . . 134–135, 142, 145, 147, 222

Lepirudin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142, 145, 147Cardiopulmonary bypass (CPB) surgery . . . . . . . . . . . . . . . 71

Lepirudin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71Cardiopulmonary bypass models . . . . . . . . . . . . . . . . . . 70–71Cardiovascular disease (CVD). . . . . . . . .194–195, 204–205,

210–211, 215, 262, 272pharmacogenomics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277

Catheter-induced thrombosis models . . . . . . . . . . . . . 64, 120Cell adhesion molecules (CAMs) . . . . . . . 85, 111, 271–272,

280, 261, 265as surrogate markers . . . . . . . . . . . . . . . . . . . 114, 262, 272

Cellular signaling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271tissue factor (TF)/activated factor VIIa . . . . . . . . . . . 124,

162, 243Cerebral blood flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213Certoparin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121

chemical characteristics . . . . . . . . . . . . . . . . . . . . . . . . . .111CFR, see Cyclic flow reductions (CFR)Chandler loop . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5–6Chemically-induced thrombosis models . . . . . . . . . . . . . . . 54Chemotaxis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117

heparin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117Clopidogrel

ACTIVE W study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184acute myocardial infarction (acute MI) . . . . . . . . . . . . . 34aspirin . . . . . . . . . . . . . . . . . . . . . . . . . . . 204–206, 208–215asymptomatic patients . . . . . . . . . . . . . . . . . . . . . . 208, 214cerebrovascular disease . . . . . . . . . . . . . . . . . 208, 212, 214clinical trials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208, 216combination therapy . . . . . . . . . . 204–206, 208–213, 215coronary artery bypass graft (CABG)

surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208, 222coronary artery disease . . . . . . . . . . . . . . . . . . . . . . . . . . 214limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208, 222non ST segment elevation. . . . . . . . . . . . . . . . . . .212, 215percutaneous coronary intervention

(PCI) . . . . . . . . . . . . . . . . . . . . . . . . . . 208, 212, 224stent thrombosis . . . . . . . . . . . . . . . . . . . . . . . 209, 224–225

ClottingCancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242–243Cancer (tumor) cell-induced . . . . . . . . . . . . 244–245, 247

indicators of . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244–245Dabigatran . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145, 172defects in . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

mouse genetic models . . . . . . . . . . . . . . . . . . . . . 82, 85Factor Xa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184–185heparin-induced thrombocytopenia (HIT) . . . . . . . . 236pharmacogenomics . . . . . . . . . . . . . . . . . . . . . . . . . 277–287TGN–167 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173–174thrombelastography (TEG) . . . . . . . . . . . . . . . . . . . . . . 3–5thrombin induced clot formation . . . . . . . . . . . . . . . 55–57

canine coronary artery . . . . . . . . . . . . . . . . . . . . . 55–57rabbit femoral artery . . . . . . . . . . . . . . . . . . . . . . 55–57

tissue factor (TF) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243in vitro assays for

formation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8lysis . . . . . . . . . . . . . . . . . . . . . . . . . . 4–6, 13–14, 64, 83strength . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

Coagulationangiogenesis . . . . . . . . . . . . 111, 114, 120, 122–125, 127,

161–162, 243, 248

ANTICOAGULANTS, ANTIPLATELETS, AND THROMBOLYTICS

Subject Index 311

animal models of . . . . . . . . . 30–33, 76, 89–90, 111, 120,162–163, 166, 185, 241

cancer . . . . . . . . . 118–119, 121, 122–124, 127, 159, 231,234, 236, 241–243, 248, 255

molecular markers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242in vitro models of . . . . . . . . . . . . . . . . . . . . 2, 5, 15, 30–33,

86, 90, 162, 164, 235Coagulation disorders . . . . . . . . . . . . . . 31–32, 110, 114, 168,

262, 268, 279mouse genetic models of . . . . . . . . . . . . . . . . . . . . 162, 247

Coagulation factor II (prothrombin) . . . . . . . . 281, 283, 285Polymorphisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281, 285

Coagulation factor IX propeptide . . . . . . . . . . . . . . . . . . . . 286mutation at ALA-l0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286

Coagulation factor V Leiden. . . . . . . . . . . . . . . . . . . .285–286R506Q. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285–286

Collagen . . . . . . . . . . . . 1, 6, 9–10, 17, 20–22, 25, 31, 37, 64,66, 71–76, 84–86, 111, 203, 215, 222,231–234, 280

Collagen-coated surfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21Preparation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .21

Copper coil-induced canine arterial thrombosis . . . . . . . . 62Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62

Coronary artery plaque . . . . . . . . . . . . . . . . . . . . . 48, 213, 272platelet aggregation and monitoring . . . . . . . 6, 8, 75–76transition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

Coronary artery thrombosis model . . . . . . . . . . . . . . . . 49, 57Canine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44, 57

Coronary blood flow . . . . . . . . . . . . . . . . . . . . . . . . . . 34–35, 41Coronary thrombosis model (Folts Model) . . . . . . 33–45, 47

mechanical-induced . . . . . . . . . . . . . . . . . . . . . . . . . . 33–45stenosis-induced . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40

Crohn’s disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116, 267CURE trial . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208,

211–212, 214Cyclic flow reductions (CFR) . . . . . . . . . . . . . . . . . . 34, 38, 91

D

Deep venous thrombosis (DVT) . . . . . . . . 91, 110, 115, 119,121, 127, 141, 158, 160–161, 163–165,167–169, 171, 173, 182, 188–194, 196,285–286

rivaroxaban . . . . . . . . . . . . . . . . . . . . . . . . . . . 160, 167–168,188–194, 196

Dextran . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54, 60, 79fluorescein-labeled . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

DiabetesClopidogrel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .208, 212Prasugrel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208, 225Warfarin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183

DIC, see Disseminated intravascular coagulation (DIC)Direct thrombin inhibitors . . . . . . . . . . . . . . 54, 71, 142–145,

158, 170–172, 236heparin-induced thrombocytopenia

(HIT) . . . . . . . . . . . 142–145, 158, 171–172, 236Disseminated intravascular coagulation

(DIC) . . . . . . . . . . . . . . . . . . 69–70, 148, 161, 242animal models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70

Drug metabolizing enzymes. . . . . . . . . . . . . . . .281–282, 287gene polymorphisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282

phase I enzymes . . . . . . . . . . . . . . . . . . . . . . . . 281–282phase II enzymes . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282

Drug transporter gene polymorphisms . . . . . . . . . . . 282–283DVT, see Deep venous thrombosis (DVT)DX-9065a . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90, 165

E

EAE, see Experimental autoimmune encephalomyelitis(EAE)

Echistatin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232–233ECM, see Extracellular matrixElderly . . . . . . . . . . . . 143, 165–166, 183, 186, 189, 196–197ELISA, see Enzyme-linked immunosorbent assay (ELISA)Embryonic lethal phenotype

factor II (prothrombin) . . . . . . . . . . . . . . . . . . . . . . . . . . . 82factor X . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83thrombomodulin. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .83

Endothelial nitric oxide synthase (eNOS) . . . . . . . . . . . . 285Enoxaparin

anticoagulant effects . . . . . . . . . . . . . . . . . . . 113, 164–165,167–168, 187, 193, 247

coronary thrombosis models . . . . . . . . . . . . . . . . . . . . . . 44heparin-induced thrombocytopenia (HIT) . . . . . . . . 148hip replacement . . . . . . . . . . . 91, 168, 170, 173, 190–191metastasis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247–249tumor cell-induced clotting . . . . . . . . . . . . . . . . . . . . . . 247compared to . . . . . . . . . . . . 113, 166, 168–169, 173, 187,

189, 191, 193, 247dabigatran . . . . . . . . . . . . . . . . . . . . . . . . . 167, 172–173LY-517717 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169razaxaban. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .165YM–150 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168

Enzyme-linked immunosorbent assay(ELISA) . . . . . . . . . 136–138, 140–141, 149, 270

antiheparin-PF4 antibody-linked . . . . . . . 140–141, 148EPIC trial . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207EPILOG trial . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207Eptifibatide (Integrilin) . . . . . . . . . . . . . . . . . 37, 91, 207, 268Erythrocytes . . . . . . . . . . . 14–17, 20, 24–25, 43, 58, 66, 293

Assays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15–18, 22Aggregation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16Deformability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

Experimental autoimmune encephalomyelitis(EAE) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266–267

Experimental thrombocytopenia orleucocytopenia . . . . . . . . . . . . . . . . . . . . 72–73, 254

Extracellular matrix (ECM) . . . . . . . . . 1, 21, 120, 261, 264,266–267

Extracorporeal thrombosis models . . . . . . . . . . . . . . . . . 71–72

F

Factor V Leiden, see Coagulation factor V LeidenFactor Xa

catalytic activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251coagulation propagation . . . . . . . . . . . . . . . . . . . . 163–170molecular target . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146

Factor Xa inhibitorsclinical trials . . . . . . . . . . . . . . . . . . . . . . . . 89, 92, 146, 175effect on coagulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147mechanism of action . . . . . . . . . . . . . . . . . . . 117, 233, 235

direct . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117, 235indirect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235

natural . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159, 161, 251oral . . . . . . . . . . . . . . . . . . . . . 142, 150, 165, 167–168, 196tissue factor pathway inhibitor (TFPI) . . . . . . . . . . . . 247

Factor XIII . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281, 285Polymorphisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281, 285

Ferric chloride-induced carotid arterythrombosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53–55

Ferric/ferrous chloride . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54


Subject Index

Fibrinogen . . . . . . . . . 1–2, 5–6, 9–10, 12–13, 34, 37–38, 57,68–69, 82–85, 87, 113, 115, 123, 143, 207,230–236, 242–247, 264–265, 270, 280

adhesion assay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270Fibrinolytic activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13, 61

euglobuliin lysis time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13Fisetin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233FIX . . . . . . . . . . . . . . . . . . . . . . . . . . . 80–81, 87, 161, 170, 286

deletion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80–81genetic model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80

Flavanoids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231–233Flow cytometry . . . . . . . . . . . . . . . . . . . . . . . . . . 17–19, 25, 139Fluid shear stress . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2, 17, 24

assays of . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17Fluorescein-labeled dextran, see DextranFolts model, see Coronary thrombosis modelFondaparinux (fondaparinux sodium)

drawbacks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164heparin-induced thrombocytopenia (HIT) . . . 142, 146,

158, 164indications and usage . . . . . . . . . . . . . . . . . . . . . . . . . . . 146metastasis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250pharmacokinetics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163venous thromboembolism (VTE) . . . . . . . 158, 163, 165and warfarin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158, 164compared to Idraparinux . . . . . . . . . . . 142, 146, 163–164

FVIIai . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161

G

Garlic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233–234platelet aggregation. . . . . . . . . . . . . . . . . . . . . . . . .233–234

Genes . . . . . . . . . . 73, 80–82, 84–85, 87, 183, 279, 281–286,291–292, 303

in hematologic, hemostatic, and thromboticdisorders . . . . . . . . . . . . . . . . . . . . . . . . . . . 278–279

Genistein . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232Genetic models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33, 79–88

thrombosis and hemostasis . . . . . . . . . . . . . . . . . . . . 79–88Genetic polymorphisms . . . . . . . . . . 208, 237, 278, 281–282

drug effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287GPIIb/IIIA antagonists . . . . . . . . . .5, 17, 22, 37, 44, 89–90,

207–208, 254, 268–269oral . . . . . . . . . . . . . . . . . . . . . . 44, 207, 216, 254, 268–269

GUSTO V trial . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210

H

Harbauer model, see Arterial and venous thrombosis model(Harbauer)

Hematological disorders . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285Pharmacogenomics . . . . . . . . . . . . . . . . . . . . . . . . . 285–286

Hemophilia B . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80–81Hemostasis . . . . . . . . 32, 33, 73, 77–89, 114, 163, 169–170,

184, 221, 230, 237, 251genetic models . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33, 77–88

Hemostatic disorders . . . . . . . . . . . . . . . . . . 32, 168, 208, 279genes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33, 279–280

Heparinanti-inflammatory . . . . . . . . . . . . 110, 115–118, 123, 126asthma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111, 116bioavailability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112–113cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . .110–111, 118–127deep vein thrombosis (DVT). . . . . . . . . . . . . . . . . . . . .110endothelial release . . . . . . . . . . . . . . . . . . . . . . . . . . 120, 134indications . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110–111, 113inflammatory bowel disease . . . . . . . . . . . . . 111, 116–117

laser-induced thrombosis models . . . . . . . . . . . . . . . . . . 58mechanisms . . . . . . . . . . . . . . . . . . . . . . . . . . . 116–119, 122metastasis . . . . . . . . . . . . . . . 110–111, 119–120, 122–127oral . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119, 121, 123pharmacology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89–90platelet–cancer cell adhesion . . . . . . . . . . . . . . . . . . . . . 120stasis thrombosis model . . . . . . . . . . . . . . . . . . . . . . . 31, 58thrombosis . . . . . . . . . . . . . . . . . . . . . . . . 33, 110, 113–118tissue factor pathway inhibitor (TFPI) . . . . 33, 110, 115

Heparin-induced thrombocytopenia (HIT)angiogenesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125–126anticoagulants . . . . . . . . . . . 136, 140, 142–144, 146–147antiheparin-PF4 antibody ELISA . . . . . . . . . . . 137, 141clinical manifestations . . . . . . . . . . . . . . . . . . . . . . 135–136flow cytometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139laboratory methods . . . . . . . . . . . . . . . . . . . . . . . . . 136–141methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .141–148pathophysiology . . . . . . . . . . . . . . . . . . . . . . . . . . . 146–147platelet aggregation assay . . . . . . . . . . . . . . . . . . . . . . . . . 90serotonin release assay (SRA) . . . . . . . . . . . . . . . . 137–138type I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136–138type II . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136–138

Heparin pentasaccharide . . . . . . . . . . . . . . . . . . . 116, 162–164vs. low molecular weight heparin

(LMWH) . . . . . . . . . . . . . . . . . 116–117, 134, 162Heparin-platelet factor 4 (PF4) . . . . . . . 133–138, 140–141,

145–146, 148–149, 163–164, 280Herbal supplements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230

antiplatelets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230drug interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . 230, 237

Hip replacement . . . . . . . . 145, 168, 170, 173, 181, 189–191venous thromboembolism (VTE) . . . . . . . . . . . . 181, 189

Hirudinheparin-induced thrombocytopenia (HIT) . . . . . . . . 135laser-induced thrombosis models . . . . . . . . . . . . . . . . . . 58PEG-hirudin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44, 63

Hirulog, see Bivalirudin (Hirulog)HIT, see Heparin-induced thrombocytopenia (HIT)Human soluble selectin assays . . . . . . . . . . . . . . . . . . 263–264Human soluble VCAM-I . . . . . . . . . . . . . . . . . . . . . . 272–273Hypercoagulable assays . . . . . . . . . . . . . . . . . . . . . . . . . . 31, 147

I

ICAM-I, see Inter-cellular adhesion molecule-1 (ICAM-1)Immobilized platelets . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22–26

cell adhesion to. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .22–26Immunoglobulin . . . . . . . . . . . . . . . . 111, 136, 236, 261–262,

265–266, 271–273Inflammation

Heparin . . . . . . . . . . . . . . . . 110, 113–118, 122, 148–150,159, 162, 295

low molecular weight heparin . . . . . . . . . . . . 44, 109–127Thrombosis . . . . . . . . . . . 40, 85, 88, 110, 113–118, 123,

147–149, 158, 162, 293Inflammatory bowel disease (IBD) . . . . . . . . . . . . . . . . . . 111,

116–117, 126, 267Injury-induced (electrolytic) arterial thrombosis

model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45–48canine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44

Innohep, see Tinzaparin (Innohep(tm))Integrilin, see Eptifibatide (Integrilin)Integrin-based assays . . . . . . . . . . . . . . . . . . . . . . . . . . 269–271Integrins

alpha 2b beta . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3alpha 4 integrins . . . . . . . . . . . . . . . . . . . . . . . . . . . 265–267


Subject Index 313

angiogenesis . . . . . . . . . . . . . . . . . . . . . . 122, 126, 262, 267bacterial infection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267beta 1 integrins . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266–267gastrointestinal disease . . . . . . . . . . . . . . . . . . . . . . . . . . 267inflammatory disease . . . . . . . . . . . . . . . . . . . . . . . 266–267integrin-based assays . . . . . . . . . . . . . . . . . . . . . . . 269–271structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265vascular endothelial growth factor (VEGF) . . . 115, 127vascular remodeling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269

Inter-cellular adhesion molecule-1 (ICAM-1) . . . . . . . . 114Assays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113

Intravascular thrombosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . .72

K

Knee replacement surgeryfondaparinux sodium. . . . . . . . . . . . . . . . . . . . . . . . . . . . 146prophylaxis . . . . . . . . . . . . . . . . . . . . . . . 146, 161, 166, 169thromboembolic events . . . . . . . . . . . . . . . . . . . . . . . . . .145tinzaparin (lnnohep R©) . . . . . . . . . . . . . . . . . . . . . 296, 302

L

Large animal thrombosis models . . . . . . . . . . . . . . . . . . . . . . 29Laser-induced thrombosis models . . . . . . . . . . . . . . . . . 57–59Lepirudin

acute venous thromboembolism (acute VTE) . . . . . . 158heparin-induced thrombocytopenia (HIT). . . . . . . .142,

144–147, 149, 158, 172Light transmittance aggregometry assay . . . . . . . . . 269–2705-lipoxygenase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86, 281Low molecular weight heparin (LMWH)

acute myocardial infarction (AMI) . . . . . . . . . . . . . . . 255cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . .110–111, 118–127chemical characteristics . . . . . . . . . . . . . . . . . . . . . 111, 113combination strategies . . . . . . . . . . . . . . . . . . . . . . . . . . 136complications of . . . . . . . . . . . . . . . . . . . . . . . . . . . 116, 125

deep vein thrombosis (DVT) . . . . . . . . 119, 121, 127deep vein thrombosis (DVT) . . 110, 115, 119, 121, 127heparin-induced thrombocytopenia (HIT) . . . . . . . . 158inflammatory disease . . . . . . . . . . . . . . . . . . . 110, 114–116metastasis . . . . . . . . . . . . . . . . . . . . . . . . 110–111, 119–120pharmacokinetics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145pharmacology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187thrombosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110, 113–118tissue factor pathway inhibitor (TFPI) . . . . . . . 110, 122,

126–127release . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112, 122, 127

venous thromboembolism (VTE) . . . . . . . . . . . . 120–121vs. unfractionated heparin

(UFH) . . . . . . . . . . . . . . . . . . . . 110, 112–113, 121

M

MDR-1 gene polymorphisms . . . . . . . . . . . . . . . . . . . . . . . 283Melagatran . . . . . . . . . . . . . . 44, 172–173, 175, 183–184, 188

see also XimelagatranMetabolism . . . . . . . . . . . . . . 37, 90, 167, 182–183, 196–197,

226, 237, 282–287, 293Pharmacogenomics . . . . . . . . . . . . . . . . . . . . . . . . .283, 285

Metastasisblood-borne. . . . . . . . . . . . . . . . . . . . . . . . . . .243–244, 252coagulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241–254experimental models . . . . . . . . . . . . . . . . . . . . . . . .243, 254heparin. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .246–251low molecular weight heparin

(LMWH). . . . . . . . . . . . . . . . . . . . . .247–250, 255

Methylenetetrahydrofolate reductase gene(MTHFR) . . . . . . . . . . . . . . . . . . . . . . . . . 285, 295

C677T polymorphism . . . . . . . . . . . . . . . . . . . . . . 285, 295Methylsulfonylmethane (MSM) . . . . . . . . . . . . . . . . . . . . . 235Microvascular thrombosis trauma models . . . . . . . . . . . . . . 70Monocytic THP-1 cell-platelet adhesion assay . . . . . . 22–24Mothers

Innohep(tm) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302Labor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303Pregnancy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302

Multidrug-resistant (MDR-l) gene polymorphisms . . . . 283Myocardial infarction

acute, see Acute myocardial infarction (AMI)low molecular weight heparin

(LMWH) . . . . . . . . . . . . . . . . . . . . . 110, 137, 142tissue factor pathway inhibitor

(TFPI) . . . . . . . . . . . . . . . . . . . . . . . . 110–113, 122unfractionated heparin (UFH). . . . . . . . . . . . . . .110, 112

N

NAPc2, see Nematode anticoagulant proteinc2 (NAPc2)

Nematode anticoagulant protein c2 (NAPc2) . . . . . . . . . 161,175, 253

Novastan, see Argatroban (Novastan)

O

Oral platelet GPIIb/III3a antagonists . . . . . . . . . . . . . . . . . . 5Orbofiban . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23, 268

P

Parallel-plate flow chamber . . . . . . . . . . . . . . . . . . . . . . . 20–24PCI, see Percutaneous coronary interventionPECAM–1, see Platelet endothelial cell adhesion

molecule-1 (PECAM-1)Percutaneous coronary intervention (PCI) . . . . . . . . 71, 143,

145, 206, 208–209, 211–212, 214–215, 224,226

PE, see Pulmonary embolism (PE)PF4 . . . . . . . . . . . . . . . . . . . . . . . 133–138, 140–141, 145–146,

148–149, 163–164, 280P-glycoprotein . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282–283Pharmacogenomics

cardiovascular disease (CVD) . . . . . . . . . . . 277, 283, 285definition of . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277hematological disorders . . . . . . . . . . . . . . . . . . . . . . . . . 285single nucleotide polymorphisms (SNPs) . . . . . . . . . . 278

drug metabolizing enzymes . . . . . . . . . . . . . . . . . . 281drug transporters . . . . . . . . . . . . . . . . . . . . . . . 282–283

warfarin therapy . . . . . . . . . . . . . . . . . . . . . . . . . . . 282–284Plaque . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48, 213, 272

coronary artery transition . . . . . . . . . . . . . . . . . 34, 49, 214Platelet aggregation assays . . . . . . . . . . . . . . . . . . . . . . . . . . . 90

laser-induced . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57–59Born method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6–9

Platelet antagonistsAssays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17endothelial cell migration . . . . . . . . . . . . . . . . . . . . . . . . 269Platelet endothelial cell adhesion molecule-1

(PECAM-1) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272in vitro thrombosis flow models . . . . . . . . . . . . . . . . . . . 17

Platelet 1251-fibrinogen-binding assay . . . . . . . . . . . . . . . 270Platelet glycoprotein Ia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286

gene polymorphisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286


Subject Index

Platelet glycoprotein IIIa . . . . . . . . . . . . . . . . . . . . . . . . . . . 286T393C gene polymorphism. . . . . . . . . . . . . . . . . . . . . .286

Platelet GPIIb/IIIA antagonists . . . . . . . . . . . 5, 51, 89, 113,207, 254, 268

chronic therapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268–269thromboembolic events . . . . . . . . . . . . . . . . . . . . . . . . . .269

Platelet perfusion studies . . . . . . . . . . . . . . . . . . . . . . . . . 21–22Platelet(s)

Adhesion . . . . . . . . . . . . . . . . . . . 6, 19–25, 48, 58, 60, 79,86, 203, 222, 232, 234

Antagonists . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16–19Assays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139, 147Isolation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21preparation

platelet-poor plasma (PPP) . . . . . . . 4, 7–8, 248, 269platelet-rich plasma (PRP) . . . . . . . . . . . . . . . . . . 6–9,

76, 137, 139, 239washed platelets (WP) . . . . . . . . . . . . . . 6–9, 137, 139

Prasugrelacute coronary syndrome (ACS) . . . . . . . . . . . . . . . . . . 222percutaneous coronary intervention (PCI) . . . . 224, 226pharmacokinetics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223renal impairment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226

PregnancyInnohep(tm) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302–303low molecular weight heparin (LMWH) . . . . . 302–303sickle cell anemia . . . . . . . . . . . . . . . . . . . . . . . . . . . 302–303venous thromboembolism (VTE) . . . . . . . . . . . . 302–303

Prothrombin, see Coagulation factor II (prothrombin)Pulmonary embolism (PE)

Acute . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115, 158, 188arterial blood gases (ABG). . . . . . . . . . . . . . . . . . . . . . . .43heparin-induced thrombocytopenia

(HIT) . . . . . . . . . . . . . . . . . . . . . . . . . 133–136, 158Tinzaparin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121unfractionated heparin (UFH). . . . . . . . . . . . . . .121, 158

Q

Quantitative sandwich immunoassay technique, seeEnzyme-linked immunosorbent assay (ELISA)

R

Rabbit femoral artery thrombosis . . . . . . . . . . . . . . . . . . . . . 91Rat jugular-vein clamping thrombosis model . . . . . . . . . . . 31Recombinant inactivated FVIIa (FVIIai) . . . . . . . . . . . . . 161Recombinant nematode anticoagulant protein c2

(NAPc2) . . . . . . . . . . . . . . . . . . . . . . . 161, 175, 253Recombinant tick anticoagulant peptide (rTAP) . . . . . . . . 51Recombinant tissue factor pathway inhibitor

(TFPI) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126–127Recombinant tissue plasminogen activator

(t-PA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57, 69, 91Renal cell carcinoma. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .123Reperfusion therapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209

acute myocardial infarction (AMI) . . . . . . . . . . 34, 39, 91Restenosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33, 47, 263, 269

animal models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31, 33Rivaroxaban

acute coronary syndromes (ACS) . . . . . . . . . . . . 194–195clearance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195–196clinical trials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208hepatotoxicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196orthopedic surgery . . . . . . . . . . . . . . . . . . . . . . . . . 166, 196

complications of, 182; venous thromboembolism(VTE) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158, 160

rNAPc2, see Recombinant nematode anticoagulant proteinc2 (NAPc2)

Rose Bengal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32, 54, 60rTAPr, see Recombinant tick anticoagulant peptide (TAP)r-TFPI, see Recombinant tissue factor pathway inhibitor

(TFPI)rt-PA, see Recombinant tissue plasminogen activator

(t-PA)

S

SelectinsClassification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281human soluble selectin . . . . . . . . . . . . . . . . . . . . . . 263–264

assay . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263–264Selenium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235Serotonin release assay (SRA). . . . . . . . . . . . . . . . . . .137–140Serum aminotransferase . . . . . . . . . . . . . . . . . . . . . . see Alanine

aminotransferaseShear-induced platelet aggregation . . . . . . . . . . . . . . 2, 17, 20Sickle cell anemia (SCA)

clinical management . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298adult . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 296child . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 296

inflammatory mediators . . . . . . . . . . . . . . . . . . . . . . . . . 293life expectancy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295pathogenesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295pregnancy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295prevalance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292

worldwide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292tinzaparin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294

Single-nucleotide polymorphisms (SNPs)cardiovascular disease (CVD) . . . . . . . . . . . . . . . . . . . . 283definition of . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 278drug metabolizing enzymes . . . . . . . . . . . . . . . . . . . . . . 281drug transporters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286hematological disorders . . . . . . . . . . . . . . . . . . . . . . . . . 285linked to adverse drug reactions . . . . . . . . . . . . . . . . . . 281mapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279–283warfarin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283–284

SK-BR-3 cell adhesion assay . . . . . . . . . . . . . . . . . . . 270–271Small animal thrombosis models, 60SNP, see Single-nucleotide polymorphismsSoluble adhesion molecules . . . . . . . . . . . . . . . . . . . . . 262, 272

as surrogate markers . . . . . . . . . . . . . . . . . . . . . . . . 262, 272SRA, see Serotonin release assay (SRA)Stasis-induced thrombosis model (Wessler

model) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67–69Stents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209, 214Streptokinase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .39, 210

use in animal models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40Synthetic pentasaccharides . . . . . . . . . . . . . . . . . . . . . . . . . 164

T

TAP, see Tick anticoagulant peptide (TAP)Tea extract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231–232TEG, see Thrombelastography (TEG)TFPI, see Tissue factor pathway inhibitor (TFPI)TF, see Tissue factor (TF)Thiopurine methyltransferase (TPMT) . . . . . 279, 281–282THP-1 cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22–24

phase-contrast photomicrograph . . . . . . . . . . . . . . . . . . 24Thrombelastography (TEG) . . . . . . . . . . . . . . . 3–5, 242, 248Thrombin-induced canine coronary/rabbit femoral artery

clot formation . . . . . . . . . . . . . . . . . . . . . . . . . 55–57


Subject Index 315

Thrombin inhibitors . . . . . . 5, 37, 44, 47, 54, 58, 66, 71, 89,142–145, 158, 162–163, 170–173, 183, 188,196, 235–236

Direct . . . . . . . .32, 54, 71, 142–145, 158, 163, 170–172,175, 188, 196, 235–236

Indirect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170–171, 235Thrombocytopenia . . . . . . . . . 72–75, 79, 84, 112, 120, 122,

133–150, 163–164, 171, 173–174, 236, 242,248–249, 251–252, 254, 269, 302

Cancer . . . . . . . . . . . . . . . . . . 120, 122, 158, 173, 242, 248heparin-induced thrombocytopenia (HIT) . . . 134–135,

138–141, 149Thromboembolic disorders . . . . . . . 110, 112, 114, 262, 268

Heparin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118–125Inflammation . . . . . . . . . . . . . . . . . . . . . . . . . 110, 113–118

Thrombolysis in myocardial infarction (TIMI)trial . . . . . . . . . . . . . . . . . . . . . . . . . . . 194, 208, 214

Thrombolytic agents . . . . . . . . . . . . . . . . . . 38–39, 51, 55, 213Thrombosis

Argatroban . . . . . . . . . . . . . . . . 88, 91, 142, 144, 147–149factor Xa inhibitors . . . . . . . . . . . . . . . . . . . . . . . . . 181–197genetic models . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33, 79–88heparin. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .109–127lepirudin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144–145low molecular weight heparin (LMWH) . . . . . 148–149malignancy . . . . . . . . . . . . . . . . . . . . . . . . . . . 162, 241–242tissue factor pathway inhibitor (TFPI) . . . . . . . . . . . . 162tissue factor (TF)/activated factor VIIa. . . . . . . . . . . .187

Thrombosis modelsanimal, see Animal thrombosis modelschemically-induced . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54

ferric chloride . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53electrically-induced . . . . . . . . . . . . . . . . . . . . . . . . . . . 43–44Folts, see Coronary thrombosis modelsHarbauer, see Arterial and venous thrombosis model

(Harbauer)laser-induced . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57photochemical-induced . . . . . . . . . . . . . . . . . . . . . . . 59–60positive controls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88rat jugular-vein clamping . . . . . . . . . . . . . . . . . . . . . . . . . 45in vitro . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30–32wire-coil induced . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60–62

Thrombotic disorders . . . . . . . . . . . . . . . . . . 32, 168, 208, 279Genes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 280

Tick anticoagulant peptide (TAP) . . . . . . . . . . . . . . . . . 51, 71Ticlopidine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47, 204, 222TIMI, see Thrombolysis in myocardial infarction (TIMI)

trialTinzaparin (Innohep(tm))

Cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121–122, 126half-life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113mechanism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122–125metastasis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122–127pregnancy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295sickle cell anemia (SCA) . . . . . . . . . . . . . . . 295, 298–299

acute painful crisis . . . . . . . . . . . . . . . . . . . . . . . . . . . 300clinical utility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268

Tirofibanclinical benefits . . . . . . . . . . . . . . . . . . . . . . . . 204, 207, 268combination therapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208heparin-induced thrombocytopenia (HIT). . . .110, 207unstable angina . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91, 110

Tissue factor pathway inhibitor (TFPI)angiogenesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162

tissue factor (TF)/activated factor VIIa . . . . . . . . 162anticoagulant effects . . . . . . . . . . . . . . . . . . . . . . . . 162, 300

biologic actions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161cancer . . . . . . . . . . . . . . . . . . . . . . . 110, 122, 127, 159, 255clinical relevance . . . . . . . . . . . . . . . . . 112, 122, 127, 161,

207, 299clotting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82, 247, 255deletion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81–82low molecular weight heparin (LMWH) . . . . . . . . . 110,

112, 122, 126–127, 162, 207, 255, 299metastasis . . . . . . . . . . . . . . . . . . . . . . . .110, 122, 126–127,

162, 247–248, 251–253recombinant TFPI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251release . . . . . . . . . . . . . . . . . . . . . . . . 33, 122, 207, 247–248restenosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207TFPI–2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .162Thrombosis . . . . . . . . . . . . . . . . . . . . . 82–84, 88, 110, 161tumor angiogenesis . . . . . . . . . . . . . . . . . . . . . . . . . 123, 243unfractionated heparin (UFH). . . . . . . . . . . . . . .247, 255

Tissue factor (TF)/activated factor VIIaAngiogenesis . . . . . . . . . . . 110, 114–115, 120, 123–127,

161–162, 243anti-TF antibodies . . . . . . . . . . . . . . . . . . . . . 161, 251, 255

as cancer therapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255cancer . . . . . . . . . . . . . . . . . . 110, 123–127, 158–159, 162,

242–243, 255coagulation . . . . . 2, 5, 31–32, 64, 69, 81, 115, 161, 207,

242–243, 281, 293, 299endotoxemia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162factor VII cofactor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2gene deletion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81inflammation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110inhibition of . . . . . . . . . . . . . . . . . . . . . . . . . . 110, 115, 126mediated hypercoagulation . . . . . . . . . . . . . . . . . . . . . . 120splice variants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242thrombosis . . . . . . . . . . . . . 110, 114–115, 123–124, 127,

158, 162, 255tissue factor pathway inhibitor (TFPI) . . . . . . . . . 33, 81,

110–115, 122, 207, 247, 281, 299tumor angiogenesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243tumor cell TF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250–251vascular endothelial growth factor

(VEGF) . . . . . . . . . . . . . . 123, 125, 127, 243, 249venous thromboembolism (VTE) . . . . . . . . . . . . . . . . 159

Tissue plasminogen activators (t-PA) . . . . . . . . . . . . 5–6, 13,39, 47, 49, 51, 56–57, 60, 69, 80, 82, 84, 87,91, 111

T-PA, see Tissue plasminogen activators (t-PA)TPMT, see Thiopurine methyltransferase (TPMT)Tumor factors . . . . . . . . . . . . . . . . . . . . . . . . 111, 120, 242–243

predicting sensitivity to anticoagulants . . . . . . . 244, 246Tumor metastasis . . . . . . . . . . . . . . . . . . . . . . . . . 110, 120, 122,

127, 243tissue factor pathway inhibitor (TFPI) . . . . . . . 110–111,

115, 122tissue factor (TF)/activated factor VIIa . . . . . . 100, 184

U

UFH, see Unfractionated heparin (UFH)Ulcerative colitis . . . . . . . . . . . . . . . . . . . . . . . . . . 116, 126, 267Ultrasound/Doppler . . . . . . . . . . . . . . . 41, 44, 46, 48, 52, 55,

60, 62, 206, 213Unfractionated heparin (UFH)

Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12pharmacology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89tissue factor pathway inhibitor (TFPI) . . . . . . . . . . . . 112vs. low molecular weight heparin (LMWH) . . . . . . . . 44


Subject Index

Unstable angina . . . . . 33–34, 39–40, 50, 91, 110, 161, 209,211–212, 224, 268, 204, 207

tissue factor pathway inhibitor (TFPI) . . . . . . . . . . . . . 33unfractionated heparin (UFH) . . . . . . . . . . . . . . . . . . . . 11

Urokinase . . . . . . . . . . . . . . . . . . . . . . . . . . 61, 80, 84, 280–281knock-out mice . . . . . . . . . . . . . . . . . . . . . . . . . . . 73, 82–88

V

Vascular cell adhesion molecule-1 (VCAM-1) . . . 111, 114,265, 272–273

Assays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272–273Vascular endothelial growth factor (VEGF) . . . . . 120, 123,

125, 127, 243, 249tissue factor pathway inhibitor (TFPI) . . . . . . . . . . . . 122tissue factor (TF). . . . . . . . . . . . . . . . . . . . . . . . . . .121–122

Vascular smooth-muscle cells (VSMC) . . . . . . 124–125, 269tissue factor pathway inhibitor (TFPI) . . . . . . . . . 33, 81,

110, 207, 247VCAM-l, see Vascular cell adhesion molecule-1

(VCAM-1)VEGF, see Vascular endothelial growth factor (VEGF)Vena-caval ligation model . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69Venous blood collection . . . . . . . . . . . . . . . . . . . . . . . . . . . 9, 18Venous thromboembolism (VTE) . . . . see also Deep venous

thrombosis (DVT); Pulmonary embolism (PE)anti-Xa inhibitors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .175investigational drugs . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159low molecular weight heparin (LMWH) . . . . . 121–122major orthopedic surgery . . . . . . . . . . . . . . . . . . . . . . . . 166

fracture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .169replacement, 166; hip, 166; knee, 166

NAPc2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .161, 175oral heparin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171pregnancy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302–303prevalence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165, 173treatment . . 120–122, 158–159, 163, 172–173, 188, 193

acute . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158, 163long-term . . . . . . . . . . . . . . . . 121, 158, 168, 174, 188

tumor associated VTE . . . . . . . . . . . . . . . . . . . . . . 124–125unfractionated heparin (UFH) . . . . . 112, 157–158, 302

Venous thrombosis . . . . . . . . . . . . . 31, 47, 61, 65, 67, 81, 84,118, 146, 163, 168–169, 181–182, 188–189,193–194, 236, 241, 255, 285

Venous thrombosis animal models, see Animal venousthrombosis models

Vessel-wall damage models . . . . . . . . . . . . . . . . 29–32, 64, 66Viscometric-flow cytometric methodology . . . . . . . . . . . . . 17Vitamin K . . . . . . . . . . . . . 121, 142, 146–147, 158, 182–185,

195, 229, 284, 286VLA–1, see Alpha 4 integrinVSMC, see Vascular smooth-muscle cells (VSMC)VTE, see Venous thromboembolism (VTE)

W

WarfarinCancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110cancer therapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183failure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172mechanism. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185metabolism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182natural products . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172pharmacogenomics . . . . . . . . . . . . . . . . 279, 283–284, 287resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208sickle cell anemia (SCA). . . . . . . . . . . . . . . . . . . . . . . . .292compared to LMWH . . . . . . . . . . . . . . . . . . 187, 193, 196

Washed platelets (WP) . . . . . . . . 6–9, 12, 18, 23, , 137, 139isolation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

Wessler model, see Stasis-induced thrombosis model(Wessler model)

Wire-coil induced thrombosis . . . . . . . . . . . . . . . . . . . . . 60–62

X

Ximelagatrantoxicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158, 167–168venous thromboembolism (VTE) . . . . . . . 157–161, 163,

165–174Xylazine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7, 41, 45, 59, 73

Documents

Anticoagulants, Antiplatelets, and Thrombolytics: Second Edition