CLINICAL CHEMISTRY AND LABORATORY MEDICINE

Transcription

1 2010 VOLUME 48 SUPPLEMENT 1 ISSN e-issn CLINICAL CHEMISTRY AND LABORATORY MEDICINE 2010 VOLUME 48 SUPPLEMENT 1 PP. S1-S128 CLINICAL CHEMISTRY AND LABORATORY MEDICINE 10TH EFCC CONTINUOUS POSTGRADUATE COURSE IN CLINICAL CHEMISTRY: NEW TRENDS IN CLASSIFICATION, DIAGNOSIS AND MANAGEMENT OF THROMBOPHILIA, OCTOBER 23-24, 2010, DUBROVNIK, CROATIA Edited by Ana-Maria Simundic, László Muszbek, Elizabeta Topic and Andrea Rita Horvath EDITOR-IN-CHIEF Mario Plebani, Padova

2

3 Clinical Chemistry and Laboratory Medicine Published in Association with the International Federation of Clinical Chemistry and Laboratory Medicine and the European Federation of Clinical Chemistry and Laboratory Medicine CCLM is the official journal of the Association of Clinical Biochemists in Ireland (ACBI), the Belgian Society of Clinical Chemistry (BVKC/SBCC), the German United Society for Clinical Chemistry and Laboratory Medicine (DGKL), the Greek Society of Clinical Chemistry-Clinical Biochemistry, the Italian Society of Clinical Biochemistry and Clinical Molecular Biology (SIBioC), the Slovenian Association for Clinical Chemistry and the Spanish Society for Clinical Biochemistry and Molecular Pathology (SEQC). Editor-in-Chief Mario Plebani Padova, Italy Associate Editors Giuseppe Lippi Reviews Editor Parma, Italy Philippe Gillery Reims, France Steven Kazmierczak Portland, USA Karl J. Lackner Mainz, Germany Bohuslav Melichar Olomouc, Czech Republic Gérard Siest Perspectives Editor Nancy, France John B. Whitfield Brisbane, Australia Editorial Board Marianne Abi Fadel Beirut, Lebanon Francisco V. Alvarez Oviedo, Spain Hassan M.E. Azzazy Cairo, Egypt Arnold von Eckardstein Zurich, Switzerland Andrea Griesmacher Innsbruck, Austria Jean-Louis Guéant Vandoeuvre-les-Nancy, France Wolfgang Herrmann Homburg/Saar, Germany Johannes J.M.L. Hoffmann Nuenen, Netherlands Herbert Hooijkaas Rotterdam, Netherlands Kiyoshi Ichihara Ube, Japan Ellis Jacobs New York, USA Naziha Kaabachi Tunis, Tunisia Jeong-Ho Kim Seoul, Korea Wolfgang Korte St. Gallen, Switzerland Christos Kroupis Athens, Greece Leslie Charles Lai Kuala Lumpur, Malaysia W.K. Christopher Lam Shatin, Hong Kong Janja Marc Ljubljana, Slovenia Eiji Miyoshi Osaka, Japan Michael Neumaier Mannheim, Germany Tomris Özben Antalya, Turkey Vladimir Palicka Hradec Králové, Czech Republic Mauro Panteghini Milan, Italy José M. Queraltó Barcelona, Spain Marileia Scartezini Curitiba, Brazil Gerd Schmitz Regensburg, Germany Hong Shang Shenyang, China Ziyu Shen Beijing, China Ana-Maria Simundic Zagreb, Croatia Bart Staels Lille, France Elizabeth Topić Zagreb, Croatia Gregory J. Tsongalis Lebanon, USA Pierre Wallemacq Brussels, Belgium Shengkai Yan Beijing, China K.T. Jerry Yeo Chicago, USA Ian S. Young Belfast, UK Managing Editor Heike Jahnke Berlin, Germany

4 Abstracted/Indexed in Indexed in Academic OneFile (Gale/Cengage Learning); Analytical Abstracts; Biochemistry & Biophysics Citation Index; Biological Abstracts; BIOSIS Previews; CAB Abstracts and Global Health; Chemical Abstracts and the CAS databases; Current Contents/Life Sciences; CSA Illustrata Natural Sciences; Elsevier BIOBASE/Current Awareness in Biological Sciences (CABS); EMBASE, the Excerpta Medica database; EMBiology; Index Medicus/MEDLINE; ISI Custom Information Services; Journal Citation Reports/Science Edition; Science Citation Index; Science Citation Index Expanded (SciSearch); and Scopus. CCLM is Published in Association with the International Federation of Clinical Chemistry and Laboratory Medicine and the European Federation of Clinical Chemistry and Laboratory Medicine. CCLM is the official journal of the Association of Clinical Biochemists in Ireland (ACBI), the Belgian Society of Clinical Chemistry (BVKC/SBCC), the German United Society for Clinical Chemistry and Laboratory Medicine (DGKL), the Greek Society of Clinical Chemistry-Clinical Biochemistry, the Italian Society of Clinical Biochemistry and Clinical Molecular Biology (SIBioC), the Slovenian Association for Clinical Chemistry and the Spanish Society for Clinical Biochemistry and Molecular Pathology (SEQC). The publisher, together with the authors and editors, has taken great pains to ensure that all information presented in this work (programs, applications, amounts, dosages, etc.) reflects the standard of knowledge at the time of publication. Despite careful manuscript preparation and proof correction, errors can nevertheless occur. Authors, editors and publisher disclaim all responsibility for any errors or omissions or liability for the results obtained from use of the information, or parts thereof, contained in this work. The citation of registered names, trade names, trademarks, etc. in this work does not imply, even in the absence of a specific statement, that such names are exempt from laws and regulations protecting trademarks etc. and therefore free for general use. ISSN e-issn CODEN CCLMFW All information regarding notes for contributors, subscriptions, Open access, back volumes and orders is available online at Responsible Editor(s) Prof. Dr. Prof. Mario Plebani, Università di Padova, Dipartimento di Medicina di Laboratorio, Azienda Ospedaliera di Padova, Via Nicolo Giustiniani 2, Padova, Italy. Tel.: +390 (0) , Fax: +390 (0) , mario.plebani@unipd.it Journal Manager Heike Jahnke, De Gruyter, Genthiner Straße 13, Berlin, Germany, Tel.: +49 (0) , Fax: +49 (0) , heike.jahnke@degruyter.com Responsible for advertisements Dietlind Makswitat, De Gruyter, Genthiner Straße 13, Berlin, Germany, Tel.: +49 (0) , Fax: +49 (0) , dietlind.makswitat@degruyter.com Imke Ridder Verlagsservice e. K., Rudolf-Diesel-Straße 10, Landsberg, Germany, Tel.: +49 (0) , Fax: +49 (0) , cclm@imke-ridder.de 2010 Walter de Gruyter GmbH & Co. KG, Berlin/New York Typesetting Compuscript Ltd., Shannon, Ireland Printing AZ Druck und Datentechnik GmbH, Kempten Printed in Germany Cover illustration Research lab Eisenhans. Permission to use the image on the cover granted by Fotolia LLC.

5 Clin Chem Lab Med 2010;48(Suppl.1):S1 S128, 2010 by Walter De Gruyter Berlin New York Contents EDITORIAL Special issue of the 10th EFCC Continuous Postgraduate Course in Clinical Chemistry: New Trends in Classification, Diagnosis and Management of Thrombophilia, October 23-24, 2010, Dubrovnik, Croatia Ana-Maria Simundic, László Muszbek, Elizabeta Topic and Andrea Rita Horvath REVIEWS Platelet physiology and antiplatelet agents Tim Thijs, Benedicte P. Nuyttens, Hans Deckmyn and Katleen Broos Hypercoagulable state, pathophysiology, classification and epidemiology Zrinka Alfirević and Igor Alfirević Diagnostic algorithm for thrombophilia screening Sandra Margetic s1 s3 s15 s27 Protein C and protein S deficiencies: similarities and differences between two brothers playing in the same game Zsuzsanna Bereczky, Kitti B. Kovács and László Muszbek s53 Antithrombin deficiency and its laboratory diagnosis László Muszbek, Zsuzsanna Bereczky, Bettina Kovács and István Komáromi s67 Factor V Leiden and FII testing in thromboembolic disorders Tadej Pajić Hyperhomocysteinemia and thrombophilia Mojca Božić -Mijovski Pediatric thrombosis Alenka Trampus-Bakija Thrombophilia screening at the right time, for the right patient, with a good reason Mojca Stegnar Methodological issues of genetic association studies Ana-Maria Simundic s79 s89 s97 s105 s115 Genetic basis of thrombosis Valeria Bafunno and Maurizio Margaglione s41 Pharmacogenetics guided anticoagulation Raute Sunder-Plassmann and Christine Mannhalter S119

6

7 Clin Chem Lab Med 2010;48(Suppl 1):S by Walter de Gruyter Berlin New York. DOI /CCLM Editorial Special issue of the 10th EFCC Continuous Postgraduate Course in Clinical Chemistry: New Trends in Classification, Diagnosis and Management of Thrombophilia, October 2010, Dubrovnik, Croatia Ana-Maria Simundic, László Muszbek, Elizabeta Topic and Andrea Rita Horvath This Special issue of the Journal is dedicated to the papers delivered by the highly esteemed speakers from the 10th EFCC Continuous Postgraduate Course in Clinical Chemistry entitled New Trends in Classification, Diagnosis and Management of Thrombophilia, which took place in Dubrovnik, Croatia on October 23 24, This postgraduate course was established by the Committee for Education and training (C-ET) of the European Federation of Clinical Chemistry and Laboratory Medicine (EFCC). During the last 9 years, it has been co-organized by the Croatian Society for Medical Biochemists, Slovenian Association for Clinical Chemistry and Inter-University Centre (IUC) in Dubrovnik. Dubrovnik courses are aimed at providing high level postgraduate education, and have covered new trends and classification and diagnosis and management of various diseases, such as cardiovascular diseases, autoimmune diseases, kidney diseases, diabetes mellitus, thyroid diseases etc. The EFCC is very proud to have this opportunity to offer such a high level educational event each year, which attracts more and more participants from all over Europe. In order to strengthen its relationship with other clinical and professional European organizations, which is one of its major strategic aims, this year the EFCC also invited the European Thrombosis Research Organization (ETRO) to join us in the organization of this 10th anniversary event. Until now, course lectures were published as whole text articles in the course Handbook, and were also made available as free full-text articles in the ejifcc at the IFCC and EFCC web sites. We are very thankful to the CCLM Editorial Board for offering us such a wonderful opportunity to publish articles from the Dubrovnik course within this special issue of the journal, which is the official journal of the EFCC. We would also like to acknowledge our special thanks to Mrs. Heike Jahnke, the Journal Managing Editor for her help and assistance in putting this issue together. The purpose of this course was to review and highlight the current understanding of the pathophysiology, epidemiology and clinical and molecular characteristics of thrombophilia. The course brought together 60 participants and 14 outstanding invited speakers. In this issue, we present reviews covering the topics of the course: the structure and function of platelets, the pathophysiology, genetic basis, classification and epidemiology of hypercoagulable states and diagnostic algorithms for thrombophilia screening. The course also covered some specific topics, such as pediatric thrombosis, deficiency of antithrombin, protein C and S, as well as the association of hyperhomocysteinemia and thrombophilia. The last session of the course focused on genetic association studies in thrombophilia research and pharmacogenetics guided anticoagulation therapy. We hope that this Special issue proves a useful and a valuable resource to all who are involved in thrombophilia research, as well as to all specialists in clinical chemistry and laboratory medicine who are involved with laboratory hemostasis on a daily basis, providing support to the care of patients. Ana-Maria Simundic 1, * László Muszbek 2 Elizabeta Topic 3 Andrea Rita Horvath 4 1 Clinical Institute of Chemistry, Emergency Laboratory Department, University Hospital SESTRE MILOSRDNICE, Zagreb, Croatia 2 Clinical Research Center, University of Debrecen, Medical and Health Science Center, Debrecen, Hungary, Past-president of the European Thrombosis Research Organization (ETRO) 3 President of the Croatian Society of Medical Biochemists, Zagreb, Croatia 4 SEALS North, Department of Clinical Chemistry, Sydney, Australia; President of the European Federation of Clinical Chemistry and Laboratory Medicine (EFCC) *Corresponding author: Ana-Maria Simundic Clinical Institute of Chemistry Emergency Laboratory Department University Hospital SESTRE MILOSRDNICE Vinogradska 29 Zagreb 1000 Croatia 2010/700

8

9 Clin Chem Lab Med 2010;48(Suppl 1):S3 S by Walter de Gruyter Berlin New York. DOI /CCLM Review Platelet physiology and antiplatelet agents Tim Thijs, Benedicte P. Nuyttens, Hans Deckmyn* and Katleen Broos Laboratory for Thrombosis Research, KU Leuven campus Kortrijk, Kortrijk, Belgium Abstract Apart from the central beneficial role platelets play in hemostasis, they are also involved in atherothrombotic diseases. Here, we review the current knowledge of platelet intracellular signal transduction pathways involved in platelet adhesion, activation, amplification of the activation signal and aggregation, as well as pathways limiting platelet aggregation. A thorough understanding of these pathways allows explanation of the mechanism of action of existing antiplatelet agents, but also helps to identify targets for novel drug development. Clin Chem Lab Med 2010;48:S3 13. Keywords: antiplatelet agents; platelet activation; platelet adhesion; platelet aggregation; platelet inhibition. Introduction Platelets play a central role in maintaining hemostasis, but are also involved in atherothrombotic diseases. Genetic polymorphisms result in large variability in platelet responsiveness to activation signals. Recently, large scale platelet transcriptome and proteome studies have been undertaken to better understand how these variations influence platelet function (1 3), as well as identification of loci for risk of myocardial infarction and coronary artery disease (4). In addition, comparative transcriptome analysis identified new platelet proteins with structural features suggesting a role in platelet function (5). In parallel, new approaches combining proteomic profiling and computational analysis will create novel insight into mechanisms of platelet activation and will in the future lead to the discovery of new protein-protein interaction network dynamics (6). Finally, all the information arising from these various approaches can be functionally validated in newly developed functional genomic models, such as the zebrafish morpholino oligonucleotide knock-out model (7, 8) or the NOD/SCID xenotransplantation model *Corresponding author: Hans Deckmyn, Laboratory for Thrombosis Research, KU Leuven campus Kortrijk, E. Sabbelaan 53, 8500 Kortrijk, Belgium Phone: q , Fax: q , Hans.deckmyn@kuleuven-kortrijk.be Received June 7, 2010; accepted September 5, 2010; previously published online November 6, 2010 (9). Taken together, these new findings will lead to a whole new era in understanding platelet function. This review focuses on the current knowledge of platelet signal transduction pathways, involving platelet adhesion, activation, aggregation and signals limiting platelet aggregation, in combination with known genetic defects and antiplatelet agents that have proven efficiency in primate models or in humans. Adhesion When the blood vessel wall is damaged, subendothelial structures are exposed to flowing blood. Collagen is the most abundant thrombogenic protein present in the subendothelial matrix. It occurs in more than 20 isoforms, with fibrillar collagen type I and III being the most abundant components present in the matrix lining the vessel wall. Platelets can bind directly to exposed collagen through two major receptors namely integrin a2b1 and glycoprotein (GP) VI (Figure 1). The initial tethering and binding of the platelet is called adhesion. However, under high shear conditions, such as that found in arterioles, neither a2b1, GP VI, nor a combination of both is sufficient to mediate platelet adhesion, which now requires the GP Ib-V-IX receptor complex, and its main ligand, von Willebrand Factor (vwf). vwf binds to exposed collagen fibers through its A3 domain, resulting in a shearmediated structural change in the molecule that allows GP Ib to bind to the vwf A1 domain (Figure 1). This is considered to be primarily an adhesive interaction characterized by a fast dissociation rate, slowing down platelets and allowing stronger bonds to be formed between collagen and the a2b1 and GP VI receptors (11). The importance of the GP Ib-vWF interaction is illustrated by the pathologic conditions occurring in the absence of either receptor or ligand, respectively, the Bernard-Soulier syndrome and von Willebrands disease (12, 13). GP Ib-V-IX The GP Ib-V-IX complex consists of four subunits: GP Iba, GP Ibb, GP IX and GP V, in a recently proposed 2:4:2:1 stoichiometry, where 2 GP Ibb (25 kda) subunits are bound to each GP Iba subunit (135 kda) by disulphide bonds. GP IX (22 kda) and GP V (82 kda) in turn are non-covalently associated with GP Ib (14). All subunits belong to the Leucine Rich Repeat protein superfamily, characterized by the presence of one or more stretches of approximately 24 amino acids containing multiple Leu residues, and flanked by conserved N- and C-terminal sequences containing disulphide bonds. 2010/332

10 S4 Thijs et al.: Normal platelet function Figure 1 Schematic representation of the main biochemical pathways involved in the formation of a platelet-dependent hemostatic plug wbased on (10)x. GP Iba is the major subunit and presents binding sites for most extracellular ligands, such as, vwf, thrombin, highmolecular weight kininogen, P-selectin and FXII. The crystal structure of GP Iba bound to the A1 domain of vwf has been resolved and suggests long-range electrostatic interactions as the major driving force, with two contact sites emerging in the final complex. One of these contact sites depends on conformational changes in the b-switch of GP Iba, thereby forming a b-hairpin and b-bimolecular sheet, with the other depending on changes in the vwf A1 domain to uncover the binding site (15, 16). Upon binding, clustering of GP Ib-V-IX receptors occurs, enabling multiple interactions between platelets and multimeric vwf molecules. The GP Ib-V-IX complex localizes to the lipid raft fraction of the cell membrane. This location is attractive for signalling molecules, allowing it to transmit weakly activating signals through a number of cytoplasmic proteins including immunoreceptor tyrosine-based activation motif (ITAM) containing proteins (17) z, another adaptor molecule, is also functionally associated with GP Ib-V-IX and participates in the inside-out activation of aiibb3. This has been demonstrated in a transgenic mouse model expressing the human GP Iba receptor, but lacking the six C-terminal residues responsible for z binding (18, 19). A recent paper by Yuan et al. (20) suggests a double role for z, inhibiting association to the cytoskeleton, while promoting vwf binding. Other molecules associated with the cytoplasmic domain of the GP Ib-V-IX complex are proteins involved in Ca 2q signalling, such as the regulatory p83 subunit of phosphoinositide-3-kinase, calmodulin and Src kinases, as well as proteins playing part in GP Iba-induced inside-out aiibb3 activation, such as the recently discovered Adhesion and Degranulation promoting Adaptor Protein (21 24). Finally, the GP Ib-V-IX complex is also linked to the cytoskeleton by its binding site for filamin, which in turn can bind to F-actin, facilitating GP Iba-mediated platelet adhesion to and translocation on a VWF matrix under high shear conditions (25). The best known inhibitor of the GP Iba-vWF interaction is the vwf A1 binding aptamer, ARC1779, which directs a dose-dependent inhibition of VWF activity and platelet function in human volunteers without causing any bleeding effects. In addition, it may serve as a new antithrombotic for the treatment of patients with acute coronary artery syndromes (26), and as a novel therapeutic for thrombotic thrombocytopenic purpura (27). Another drug candidate targeting the GP Iba binding site on vwf is the bivalent humanized Nanobody ALX This compound inhibits thrombus formation in human patients with stable angina undergoing percutaneous coronary intervention (28). Also, GP Ib-binding monoclonal antibody Fab fragments wreviewed by (29)x, such as 6B4 Fab, prevent platelet adhesion and thrombus formation in vivo without increasing bleeding time (30). Platelet collagen receptors Although platelets possess a number of receptors capable of interacting with collagen, such as CD36 and GP V, almost

11 Thijs et al.: Normal platelet function S5 all attention has focused on GP VI and a2b1. GP VI (63 kda) is part of the immunoglobulin superfamily and relies strictly on its association with the g subunit of the Fc- Receptor (FcR) in order to be successfully expressed on the platelet surface (31). In addition, the presence of FcRg is also important for triggering signalling events. Upon interaction with collagen, the cytoplasmic tail of GP VI is able to bind the SH3 domain of Fyn and Lyn, both members of the Src tyrosine kinase family. Fyn and Lyn subsequently phosphorylate Tyr residues present in the ITAM motif of FcRg. This phosphorylation allows binding of the tyrosine kinase Syk, which in turn induces a signalling cascade in a so-called signalosome, resulting in the activation of phosphoslipase Cg2 (PLCg2) and formation of second messengers and Ca 2q release from internal stores (32 34). The a2b1 collagen receptor, part of the integrin family of proteins, consists of an a2 subunit (165 kda) and b1 subunit (130 kda). It has a dependency on Mg 2q for collagen binding that increases with higher shear rates (35, 36). In resting platelets, a2b1 is present in a low-affinity state, which upon activation is changed to a high-affinity conformation, a characteristic shared by all integrins (37). Unlike most other integrins, a2b1 is not able to recognize the RGD sequence present in adhesive proteins, such as fibronectin and fibrinogen (35). By using a number of synthetic collagen-related peptides, the GFOGER (Oshydroxyproline) sequence in collagen type I was identified as the main a2b1 binding site. Other related sequences, generally designated GxOGER, that are present in collagen have also been implicated in a2b1 binding, although to a lesser extent (38 42). Signalling through a2b1 is associated with Src tyrosine kinases, Syk, and further downstream with PLCg2. Thus, although both collagen receptors belong to distinct protein families, they seem to share at least part of their outside-in signalling cascade. Recent research from our group suggests two other cytosolic proteins, RhoGDI-b and Slap-2, that interact with the a2 and b1 tail (43). Despite extensive research on both collagen receptors over the years, and the fact that they share at least part of their signalling molecules, the exact interplay between the two has long been a topic of discussion. This debate arises mainly because of conflicting results obtained by different groups, which can be ascribed partially to methodological differences and the difference between human and murine mechanisms. Based on results obtained in human platelets, the original 2- site, 2-step model was proposed where a2b1 is important for establishing a high-affinity bond with collagen, followed by a lower-affinity bond with GP VI which then would initiate the intracellular signal transduction (11). This model was questioned when it was found that a2b1 can induce platelet activation via the Src kinase signalling cascade (44), and therefore, a new unifying model was proposed (45). In this model both GP VI and a2b1 are important for mediating stable adhesion, with the specific contribution of both depending on individual platelet responsiveness and the amount and nature of exposed collagen. A recent paper by Pugh et al. (46) might provide the answer regarding the exact role of the two collagen receptors and the GP Ib-V-IX VWF interaction. Using the synthetic peptides CRP-XL, GFOGER and VWF-III, specific for GP VI, a2b1 and GP Ib-V-IX, respectively, the authors managed to study individually the effects of the three main adhesionreceptors. They conclude that a2b1 and GP Ib-V-IX can both support substantial platelet activation and thrombus formation on collagenous substrates, whereby the relative importance of the GP Ib-VWF interaction increases at higher shear rates. However, even the combined action of a2b1 and GP Ib-V-IX is insufficient to promote full aiibb3 activation and requires the strong activating signals generated by GP VI, which contributes little to platelet adhesion itself at all shear rates (Figure 1). Due to their shared signalling molecules and their importance in platelet adhesion, especially under low shear conditions, GP VI an a2b1 can both serve as suitable drug targets (47). Most of the work has been conducted on GP VI, which is not surprising given its important role in generating activation signals (46). A successful strategy has been the development of inhibitory monoclonal antibodies such as OM2 or the 9O12.2 Fab. These have antithrombotic effects in baboons, only slightly influencing the bleeding time (48, 49). In addition, it has been shown that the angiotensin II type I receptor-antagonist Losartan, a drug now prescribed to patients with high blood pressure, and its active metabolite EXP3179, reduced platelet adhesion in a mouse carotid artery injury model by directly interacting with the N-terminal Ig-like domain of GP VI (50, 51). Amplification of the platelet activation Phospholipase C, Ca 2H and granule release Binding of platelets through its direct collagen receptors triggers tyrosine kinase activity as described above, and induces the platelet activation cascade. In this process, the tyrosine phosphorylated PLCg, and especially PLCg2, plays a central role. PLCg hydrolyses its substrate phosphatidylinositol 4,5- bisphosphate (PtdIns4,5P2) yielding water soluble inositol 1,4,5 trisphosphate wins(1,4,5)p 3 x and membrane associated diacylglycerol (DAG). Ins(1,4,5)P 3 binds and opens intracellular Ca 2q channels on calcium storage sites within the platelets, known as the dense tubular system. This allows an influx of free calcium into the cytoplasm. Depletion of Ca 2q concentrations in the storage sites activates store-operated calcium entry in platelets by releasing the sensor STIM1 from the endoplasmic reticulum into the cytosol. This then further translocates to interact with plasma membrane calcium channels, the most prominent in platelets being Orai1 or CRAC channel moiety 1 (52, 53). This results in replenishment of Ca 2q from the surrounding medium into the platelet through capacitative Ca 2q entry. The increase in cytosolic Ca 2q concentrations allows numerous Ca 2q dependent processes to begin or to accelerate. Alternatively, membrane bound DAG functions together with Ca 2q, bound to phosphatidylserine, as an internal recep-

12 S6 Thijs et al.: Normal platelet function tor for the serine/threonine protein kinase C (PKC). As a consequence, PKC translocates from the cytosol to the membrane to become activated. Activated PKC phosphorylates a variety of substrates, resulting primarily in platelet stimulatory activity, and in particular in platelet secretion of a- and dense-granules. This results in the release of proteins, such as fibrinogen, vwf, thrombospondin-1, factor V, GAS6, plasminogen activator inhibitor 1, and platelet derived growth factors from the a-granules, and ADP, ATP, serotonin (or 5-hydroxytryptamine), Ca 2q and polyphosphates from the dense granules (54). Contradictory, PKC is also involved in some inhibitory responses, such as for example phosphorylation of the thromboxane A 2 (TXA 2 ) TP receptor that results in receptor desensitisation (55). Apart from its main substrate pleckstrin, PKC also phosphorylates myristoylated alanine-rich C kinase substrate. Upon phosphorylation, the latter loses its ability to crosslink actin filaments, thus facilitating secretion of the granular contents (56). In addition, phosphorylation of syntaxin 4 and platelet seci protein by PKC relieves their inhibitory action on the formation of Soluble N-ethylmaleimide-sensitive fusion protein Attachment protein Receptorcomplexes (SNARE) which are involved in membrane fusion and are necessary to mediate the exocytosis of a-granules (57). Taken together, PKC activation stimulates platelets to secrete agonists, such as ADP, ATP, serotonine and GAS6 through several mechanisms. Together with TXA 2, this leads to amplification of platelet recruitment. Phospholipase A 2 and amplification via TXA 2 Phospholipase A 2 (PLA 2 ), a key enzyme in the formation of arachidonic acid (AA) and TXA 2, is activated by increased concentrations of cytosolic Ca 2q. Platelets contain secreted (s)pla 2, localized in a-granules, and cytosolic (c)pla 2 -a. However, only (c)pla 2 -a is activated upon agonist stimulation. Activated (c)pla 2 -a cleaves platelet glycerolphosholipids, such as phosphatidylcholine, phosphatidylethanolamine and phosphatidylinositol at their sn-2 position to release AA. (c)pla 2 -a contains a C2 domain at its N-terminus which is a Ca 2q ligand binding domain and is involved in the transfer of (c)pla 2 -a from the cytosol to the membrane in response to increased concentrations of cytosolic Ca 2q. Ca 2q and DAG-regulated guanine nucleotide exchange factor I (CalDAG-GEFI) has been identified as a link between increased cytosolic Ca 2q and activation of (c)pla 2 -a (Figure 1). CalDAG-GEFI responds to increased concentrations of cytosolic Ca 2q and activates Rap1, a small GTP binding protein of the Ras family. It has been suggested that Rap1 in turn activates extracellular signal-regulated kinases (ERK) which have been implicated in the phosphorylation of (c)pla 2 -a (58). Phosphorylation of (c)pla 2 -a is important for efficient AA release. However, high concentrations of Ca 2q can overcome the need for phosphorylation (59). After its release from the membrane, AA is transformed into the unstable prostaglandin (PG)H 2 by cyclooxygenase 1 (COX-1), and finally isomerized into TXA 2 by thromboxane synthase. Once formed, TXA 2 can diffuse across the plasma membrane and activate other platelets by interacting with the TP receptor. The TP receptor is known to be a G-protein coupled receptor (GPCR), communicating with G 12 /G 13 and G q. Platelets lacking G 13 and G q are unresponsive to TXA 2, indicating their role in platelet activation (60). Stimulation of G q will activate PLCb, resulting in accumulation of Ins(1,4,5)P 3 and DAG, which in turn activate Ca 2q release and PKC. Changes in platelet shape are dependent on G 12 / G 13 stimulation, activating Rho-mediated signalling (61). The short half life of TXA 2 in solution (30 s) limits the spread of platelet activation to the original site of injury. Cyclooxygenase is the target enzyme for aspirin or acetylsalicylic acid. Aspirin covalently acetylates Ser530 on COX (62), which in the anucleated platelet results in an irreversible inhibition of the enzyme for the life time of the platelet (8 10 days), in contrast to the endothelial cells where prostacyclin (PGI 2 ) production recovers much more rapidly. In addition, low dose aspirin is also believed to spare PGI 2 production, while inhibiting platelet thromboxane formation: a first passage through the liver results in deacetylation of the low concentrations of aspirin such that the endothelium in the systemic circulation will not be exposed to aspirin, in contrast to the circulating platelets that are inhibited during their passage through the portal vein. Whatever the role of PGI 2 in the aspirin story may be, it is clear that daily intake of a low dose of aspirin is highly effective in preventing secondary thrombotic events resulting in an overall 25% reduction of the risk of thrombotic events in patients with confirmed disease of the coronary, cerebrovascular, or peripheral artery beds (63). Platelet activation via G-protein coupled receptors Almost all soluble agonists induce platelet activation by interacting with GPCRs, such as the TP-receptor discussed above. The only exceptions known to date are ATP, that activates the Ca 2q -channel P2X 1 resulting in a direct Ca 2q - influx into the platelet cytosol and GAS6 that interacts with its Tyr-kinase receptors Axl, Mer, and Sky (64, 65). All GPCRs, except for the protease activated receptor (PAR), reversibly bind their ligands and hence the effects induced depend on the dose of ligand present. Upon ligand or tethered ligand binding, the activated GPCR interacts with intracellular G-proteins, that will activate their effector molecule (66). Platelet GPCRs couple to a variety of G-proteins namely members of the G q -family that activate PLCb 2 /b 3,ofthe G i -family that inhibit adenylyl cyclase and/or activate MAPkinases, or members of the G 12 /G 13 -family that signal to small GTPases controlling actin polymerisation. All these G- protein signalling cascades can act as positive feedbackmechanisms thus ensuring further platelet (integrin) activation, shape change, granule secretion and recruitment of platelets in the growing thrombus (67). The soluble platelet agonist ADP binds to two GPCRs, P2Y 1 and P2Y 12, both required to ensure a full plateletresponse, but each coupled to its own G-protein. P2Y 1 is coupled to G q and involved in intracellular Ca 2q release,

13 Thijs et al.: Normal platelet function S7 whereas P2Y 12 is coupled to G i -type proteins (68, 69) (Figure 1). Patients lacking this receptor are suffering from a mild form of hemorragia (70). A number of compounds that target the ADP-receptor P2Y12 are now in clinical practice with clopidogrel as currently most used. Clopidogrel is a prodrug that first needs to be converted to its active metabolite by liver cytochrome P450 (CYP)-dependent oxidation. The active metabolite forms a disulfide bond with the P2Y 12 receptor, hence, like aspirin, irreversibly covalently modifying the receptor. However, due to the conversion that is needed, clopidogrel has a rather slow onset of action and is sensitive to genetic variations in CYP (so-called drug resistance), therefore, newer compounds with better characteristics are being developed, such as prasugrel (still a prodrug) and products like, ticagrelor and cangrelor (both reversible direct-acting P2Y 12 inhibitors) (70). The platelet agonist thrombin binds to the PAR, PAR1 and PAR4 at respectively low and high thrombin concentrations (Figure 1) (71). These receptors couple to a number of G- proteins, most notably G 12 /G 13 and G q and in some cases also G i. PAR receptors are activated when an extended N- terminal part of the receptor is being cleaved at a consensus sequence by their ligand. By this, a new N-terminus is exposed that folds back into the central part of the receptor and serves as a tethered ligand that induces receptor activation (72). As the cleaved PAR-receptor thus is expected to be continuously active, it is clear that in order to control PAR-dependent activation, turnoff signals are required, next to the classical ones, such as receptor desensitization, endocytosis and downregulation. Indeed evidence has been obtained that activated PAR-receptors, much like b-adrenergic receptors, become phosphorylated and inactivated by the b-adrenergic receptor kinase-2 (73). Finally, as the affinity of thrombin for the PARs clearly is lower than for the GP Ib-IX-V complex, there also is evidence that initial binding of thrombin to GP Ib might facilitate the consequent cleavage of PAR1 (74). SCH (vorapaxar) is an orally administered antiplatelet agent that attenuates thrombin induced platelet aggregation through inhibition of the PAR1 receptor without increasing bleeding risk. Phase II clinical trials indicate that this drug may be useful in addition to standard-of-care therapy in patients with coronary disease (75). Some other G q -coupled receptors that are present on platelets are 5HT 2A for serotonin, the PAF-1-receptor for PAFacether and the V-1 receptor for vasopressin (76). In platelets, both the a-subunit of G q (Ga q ) and bg-subunits efficiently activate PLCb 3, whereas Ga q -subunits do not activate PLCb 2 (77, 78). Activation of PLCbs again results in Ins(1,4,5)P 3 and DAG-formation and joins in with the activation pathways already described as a consequence of PLCg phosphorylation-dependent activation (see Phos- pholipase C, Ca 2q and granule release) ultimately leading to integrin a IIb b 3 activation (Figure 1). PAR and TP receptors can also couple to G 12 /G 13 which activates a Rho/Rho-kinase-mediated pathway resulting in the phosphorylation of the myosin light chain and therefore, contributes to platelet shape change (79). However, it has also been shown that upon blocking of G 12 /G 13, shape change can still occur by signalling through G q when sufficiently high amounts of agonist are present. G i coupled receptors, such as P2Y 12 or the a 2A -adrenergic receptor for epinephrine, can act in a different way. It has been shown that G i2, which has the highest expression levels of any G i family member in platelets, can bind to adenylyl cyclase thereby inhibiting its function. This leads to a reduced production of camp, the main inhibitory signalling molecule in platelets (see 4) and thus contributes to activation (Figure 1). To further support this activating role, it has also been shown that platelets lacking Ga i2 show reduced aggregation in response to ADP, thrombin and epinephrine (80). Not all GPCR-induced signalling is activating however, PGI 2 for instance, one of the major endothelium-derived inhibitors of platelet activation, has signals through a G s protein upon binding to its IP receptor, leading to activation of adenylyl cyclase and an increase in camp (see Signal transduction to limit platelet activation) (Figure 1) (81). Platelet procoagulant formation Platelet activation and thrombin generation, through the coagulation cascade, are intimately linked processes. Activated platelets can initiate and propagate coagulation on their surface, adequate thrombin formation is not possible without activated platelets. However, thrombin is a strong platelet activator, able to activate platelets at concentrations as low as 0.1 nm. The platelet response to thrombin is largely mediated by PAR1 and PAR4, but signaling through other receptors on the platelet surface, such as GP Ib, can play an accessory role. In platelet membranes, phospholipids are not evenly distributed between the inner and outer leaflet. In resting platelets, phosphatidylserine and phosphatidylethanolamine, negatively charged phospholipids, are found at the inside of the membrane. Upon activation, platelet membranes undergo changes by which more of these negatively charged phospholipids appear on the outside of the membrane (82). These negatively charged phospholipids then complex Ca 2q, forming a bridge with coagulation factors, such as factor Xa and factor Va. Binding of coagulation factors on the platelet membrane increases their local concentration which facilitates interaction and accelerates thrombin formation. Finally, this results in further thrombin-induced platelet activation, fibrin formation and coagulation. Platelet aggregation and integrin a IIb b 3 Platelet adhesion and activation are steps leading to the end point of the signaling cascade, platelet aggregation. In this

14 S8 Thijs et al.: Normal platelet function final step, the integrin a IIb b 3 shifts from a low affinity state to a high affinity state and efficiently binds the symmetrical fibrinogen. This step results in cross-linking of the platelets and formation of a firm platelet aggregate (Figure 1). Integrin a IIb b 3, or GP IIb-IIIa, is the only integrin expressed uniquely on platelets. With 50,000 80,000 copies per platelet, it is the major platelet integrin receptor mediating both adhesion and bidirectional signaling. Fibrinogen is the main ligand of aiibb 3, other ligands include VWF, fibronectin and vitronectin. The absence or deficiency of a IIb b 3 leads to Glanzmann thrombasthenia, characterized by a tendency for severe bleeding and a lack of platelet aggregation in response to all physiological agonists. The shift of integrin a IIb b 3 from a low to a high affinity state is considered the final common pathway of platelet activation. Therefore, the molecular machinery regulating this process represents a potential target for antithrombotic agents and has been studied intensively. Integrins are ab heterodimeric cell surface receptors consisting of a ligand binding extracellular domain sitting on two membrane-spanning legs. Ligand binding can trigger signal transduction into the cell (outside-in signaling). Integrin tail binding proteins can also induce conformational changes that alter the affinity for their ligands (inside-out signaling). Next to conformational changes that increase ligand affinity, activation can also occur by increased expression and/or clustering of integrins at the cell surface (83). The structural changes that occur when a IIb b 3 is activated have been nicely reviewed by Arnaout and co-workers (84). On the basis of structural studies, a model has been proposed that integrins are in a low-affinity state when their extracellular domains are bent. Upon activation, the integrins switch to a high-affinity state with their extracellular domains extended upwards. The exact conformational changes that occur when integrins become activated remain unclear. The switchblade model suggests that in response to an insideout signal, the integrin snaps from an inactive bent structure to an active straight conformation, analogous to the opening of a pocket knife. The deadbolt model, however, predicts that a defined region locks the integrin in an inactive state; upon activation the deadbolt slides away rendering the integrin ligand competent. In both models, these structural rearrangements finally result in a gain in ligand affinity. To date, it is still not clear how platelet agonists, acting through PLC-mediated increased Ca 2q concentrations and/or PKC-activation, results in a IIb b 3 activation (Figure 1). Although more than 20 proteins have been identified that interact with the cytoplasmic tails of a IIb b 3, the functional significance is understood for only a few. The cytoplasmic tail of the integrin b-subunit contains two well-defined motifs which serve as recognition sequences for phosphotyrosine binding (PTB) proteins, including talin and the kindlins (85). The talin binding site on the b-tail was mapped to the membrane proximal NPxY PTB protein binding motif. However, kindlins bind to the membrane distal to the NxxP PTB binding motif (86, 87). Talin is known as a key integrin activator. Binding of talin to the cytoplasmic tail results in a disruption of the salt bridge between the two integrin tails, with subsequent activation of a IIb b 3 (88). How talin is activated is not clear, but it has been proposed that it binds to the lipid second messenger PtdIns4,5P2 (89). Also, Rap1 has been implicated in recruitment of talin to the integrin tails. Rap1 induces the formation of an integrin activation complex consisting of the Rap1-GTP-interacting adaptor molecule which binds directly with Rap1-GTP and talin (90). Although abundant evidence supports the essential role of talin in activation of integrin, recent work shows that talin alone is not sufficient. Kindlin-3 deletion prevents platelet integrins to bind to their ligands. In addition, platelet aggregation is defective despite normal expression of talin (91, 92). Platelets deficient in talin show the same phenotype, suggesting that both proteins are required for integrin activation (93). How kindlins and talin cooperate to regulate integrin activation remains unclear. It is possible that they bind simultaneously or sequentially to the b-tail causing integrin activation (94). The integrin a IIb b 3 receptor recognizes its ligands through RGD motifs that are localized at the distal ends of dimeric fibrinogen (95). The binding of fibrinogen to a IIb b 3 triggers a number of signaling events that promote cytoskeletal changes, which in turn lead to spreading and stabilization of platelet thrombi. During this outside-in signaling process kindlins again seem to play a role as they bind to integrin linked kinase and the filamin binding protein migfilin, which link kindlins indirectly to the cytoskeleton (96 98). Upon ligand binding, a IIb b 3 activates non-receptor tyrosine kinases, such as Src and Syk which contribute to the stability of thrombi in vivo. Ephrin kinase A4 (EphA4), a transmembrane protein expressed on the platelet surface, is constitutively associated with a IIb b 3. When platelets are activated the expression of this protein is increased. Binding of EphA4 to its ligand, ephrinb1, facilitates tyrosine phosphorylation of the b 3 -tail required for stable platelet aggregation (99). Since a IIb b 3 is only expressed on platelets and binding of a IIb b 3 with fibrinogen is the final major step in platelet aggregation, GP IIb-IIIa antagonists are expected to have a strong and specific effect. The GP IIb-IIIa antagonists that are available clinically include abciximab, tirofiban and eptifibatide. Abciximab is a chimeric Fab fragment of the murine anti-human a IIb b 3 monoclonal antibody that blocks ligand binding to a IIb b 3 through steric hindrance. Tirofiban is a specific non-peptide antagonist of GP IIb- IIIa while eptifibatide is a cyclic heptapeptide derived from a protein found in the venom of the southeastern pygmy rattlesnake. The main mechanism of action of these drugs is the inhibition of fibrinogen binding to GP IIb-IIIa, thereby preventing platelet aggregation. Next to aspirin and clopidogrel, GP IIb-IIIa antagonists form an important component of the pharmacological management of patients undergoing percutaneous coronary intervention, such as patients with coronary artery disease (100). One practical difference between abciximab and the two small molecule agents, tirofiban and eptifibatide, is that abciximab has a longer duration of action (101). The

15 Thijs et al.: Normal platelet function S9 occurrence of minor bleeding complications and thrombocytopenia are side effects of GP IIb-IIIa antagonists. Drug induced thrombocytopenia often appears suddenly and can cause major bleeding and death. Severe thrombocytopenia is seen in 0.1% 2% of patients treated with GP IIb-IIIa antagonists within several hours after first exposure; up to 12% of patients show thrombocytopenia following a second exposure (102). Signal transduction to limit platelet activation The signal amplification and platelet recruitment mechanisms require tight control in order to limit platelet aggregate formation to the place where it is needed. Therefore, endothelial cells produce PGI 2 and nitric oxide (NO), which, respectively, increase platelet camp and cgmp concentrations, the main secondary messengers involved in platelet inhibition (Figure 1). Vascular endothelial cells produce PGI 2, which upon diffusion, binds platelet PGI 2 receptors (IP) which are coupled to adenylyl cyclase stimulating G s -protein. Adenylyl cyclase catalyses the conversion of ATP into camp, a potent inhibitor of platelet activation (Figure 1). Mutations in the PGI 2 receptor may lead to a deficiency in PGI 2 signaling and contribute to atherothrombosis (103). Also, other PGs, such as PGD 2 (via DP), PGE 2 and PGE 1 (via EP2 and EP4), and adenosine (via A2A) couple to G s -adenylyl cyclase. Increased camp concentrations result in activation of the camp-dependent protein kinase, or PKA. However, Gambaryan et al. recently showed that camp-independent mechanisms may activate PKA (104). PKA-mediated phosphorylation of the vasodilator-stimulated phosphoprotein (VASP) inhibits platelet aggregation. Aggregation and spreading of platelets is accompanied by changes in platelet morphology and in reorganization of the actin cytoskeleton, a process in which VASP is one of the key regulators. In addition, phosphorylation of VASP also inhibits binding of fibrinogen to GP IIb-IIIa (105). In 2006, Pula and co-workers, reported that PKCd also has a negative signaling role in regulating platelet aggregation through phosphorylation of VASP (106). PKA further phosphorylates the myosin light chain kinase (MLCK), impairing its binding to calmodulin and consequently reducing MLCK activity, resulting in reduced phosphorylation of myosin light chain which hampers cytoskeletal reorganisation (107, 108). Although no genetic disorders associated with VASP or MLCK have been described, defects in key proteins of the cytoskeleton organization, as found in Wiskott-Aldrich Syndrome, result in severely impaired platelet function. Synergistically to cgmp, camp inhibits Ca 2q mobilization from intracellular stores, without any major effects on the ADP-regulated cation channel (109). Nitric oxide is produced primarily in endothelial cells from L-arginine by NO synthase. Recently, however, the role of platelet produced NO in regulating platelet activation has also drawn attention wreviewed in (110)x. After diffusion of endothelial-produced NO through the platelet membrane, NO stimulates soluble guanylyl cyclase causing an increase in cgmp and concomitant activation of cgmp-dependent protein kinase (PKG). PKG inhibits Ins(1,4,5)P 3 -stimulated Ca 2q release from the sarcoplasmic reticulum and promotes ATPase-dependent refilling of intraplatelet Ca 2q stores, thereby decreasing the amount of cytosolic Ca 2q (111, 112). In addition, PKG inhibits TXA 2 receptor function through phosphorylation, thus blocking platelet activation and aggregation (113). The cgmp that is produced, in turn, might prevent platelet activation by indirectly increasing intracellular camp through inhibition of phosphodiesterase type 3 and inhibiting the activation of phospatidylinositol 3-kinase (114, 115). In addition, VASP also becomes phosphorylated following increases in cgmp by activation of guanylyl cyclase (116). Reduced responsiveness of platelets to NO, called platelet NO resistance, is involved in symptomatic ischemia, chronic heart failure and other risk factors for cardiovascular disease, and might be due to decreased bio-availability of NO or to disturbances in the integrity of the NO/ cgmp signaling pathway. Pharmacological strategies that may enhance platelet responsiveness to NO have been reviewed by Rajendran and Chirkov (117). Perspectives Recent comparative transcriptome analyses of in vitro differentiated megakaryocytes and erythroblasts have led to the identification of new platelet proteins with structural features that suggest a role in platelet function (6). In order to functionally characterize these newly identified proteins in platelet signaling pathways, there is a need for functional genomic approaches in new, previously unexplored model organisms. O Connor et al. (7) recently demonstrated the suitability of the zebrafish, Danio rerio, for functional analysis of novel platelet genes in vivo by reverse genetics in a laser-induced thrombosis model. In this model, the relatively easy knockdown in gene function following the injection of antisense morpholino oligonucleotides allows for screening of large populations of fish lacking one or more platelet receptors or signaling molecules. In a parallel approach, xenotransplantation models have been optimized allowing the in vivo investigation of human platelet function and thrombopoiesis in NOD/SCID mice (9). This model can be further explored to functionally study genetically modified platelets or new pharmacologic strategies. All the aforementioned findings discussed in this review have resulted in a significant increase in our knowledge of platelet function over the past decade. However, there still is no unifying model for the prediction or treatment of cardiovascular diseases available at the present time. However, thanks to recent large scale platelet transcriptome studies, soon we will be able to associate many of the gene polymorphisms in platelet surface receptors or signaling proteins with changes in the risk for atherothrombotic events. Fur-

16 S10 Thijs et al.: Normal platelet function thermore, it is anticipated that the development of low-cost high-throughput genotyping techniques will allow clinical application and individualization of therapies, although there is still ongoing discussions regarding the benefits and limitations of this approach (118). Platelet pharmacogenomics has nevertheless already started to link this genetic information to inter-individual variability in drug response, and as reviewed by Zuern and co-workers (119), several genetic polymorphisms can now be associated with aspirin resistance or explain the variability in treatment response to clopidogrel or abciximab. Conflict of interest statement Authors conflict of interest disclosure: The authors stated that there are no conflicts of interest regarding the publication of this article. Research funding played no role in the study design; in the collection, analysis, and interpretation of data; in the writing of the report; or in the decision to submit the report for publication. Research funding: Research funded by the Fund for Scientific Research (FWO G ). TT and BPN are funded by a fellowship from the Agency for Innovation by Science and Technology in Flanders (IWT). Employment or leadership: None declared. Honorarium: None declared. References 1. Jones CI, Bray S, Garner SF, Stephens J, de Bono B, Angenent WG, et al. A functional genomics approach reveals novel quantitative trait loci associated with platelet signaling pathways. Blood 2009;114: Macaulay IC, Carr P, Gusnanto A, Ouwehand WH, Fitzgerald D, Watkins NA. Platelet genomics and proteomics in human health and disease. J Clin Invest 2005;115: Watkins NA, Gusnanto A, de Bono B, De S, Miranda-Saavedra D, Hardie DL, et al. A HaemAtlas: characterizing gene expression in differentiated human blood cells. Blood 2009;113:e Ouwehand WH. Platelet genomics and the risk of atherothrombosis. J Thromb Haemost 2007;5(Suppl 1): Macaulay IC, Tijssen MR, Thijssen-Timmer DC, Gusnanto A, Steward M, Burns P, et al. Comparative gene expression profiling of in vitro differentiated megakaryocytes and erythroblasts identifies novel activatory and inhibitory platelet membrane proteins. Blood 2007;109: Qureshi AH, Chaoji V, Maiguel D, Faridi MH, Barth CJ, Salem SM, et al. Proteomic and phospho-proteomic profile of human platelets in basal, resting state: insights into integrin signaling. PLoS One 2009;4:e O Connor MN, Salles II, Cvejic A, Watkins NA, Walker A, Garner SF, et al. Functional genomics in zebrafish permits rapid characterization of novel platelet membrane proteins. Blood 2009;113: Lang MR, Gihr G, Gawaz MP, Muller II. Haemostasis in Danio rerio: is the zebrafish a useful model for platelet research? J Thromb Haemost 2010;8: Epub 2010 Feb Salles II, Thijs T, Brunaud C, De Meyer SF, Thys J, Vanhoorelbeke K, et al. Human platelets produced in NOD/SCID mice upon transplantation of human cord blood CD34 q cells are functionally active in an ex vivo flow model of thrombosis. Blood 2009;114: Deckmyn H, Vanhoorelbeke K, Ulrichts H, Schoolmeesters A, Staelens S, De Meyer S. Amplification loops and signal transduction pathways in platelets. In: Arnout J, de Gaetano G, Hoylaerts MF, Peerlinck K, Van Geet C, Verhaeghe R, editors. Thrombosis Fundamental and Clinical Aspects. Leuven: Leuven University Press, 2003: Savage B, Almus-Jacobs F, Ruggeri ZM. Specific synergy of multiple substrate-receptor interactions in platelet thrombus formation under flow. Cell 1998;94: Salles II, Feys HB, Iserbyt BF, De Meyer SF, Vanhoorelbeke K, Deckmyn H. Inherited traits affecting platelet function. Blood Rev 2008;22: De Meyer SF, Deckmyn H, Vanhoorelbeke K. von Willebrand factor to the rescue. Blood 2009;113: Luo SZ, Mo X, Afshar-Kharghan V, Srinivasan S, Lopez JA, Li R. Glycoprotein Ibalpha forms disulfide bonds with 2 glycoprotein Ibbeta subunits in the resting platelet. Blood 2007; 109: Dumas JJ, Kumar R, McDonagh T, Sullivan F, Stahl ML, Somers WS, et al. Crystal structure of the wild-type von Willebrand factor A1-glycoprotein Ibalpha complex reveals conformation differences with a complex bearing von Willebrand disease mutations. J Biol Chem 2004;279: Huizinga EG, Tsuji S, Romijn RA, Schiphorst ME, de Groot PG, Sixma JJ, et al. Structures of glycoprotein Ibalpha and its complex with von Willebrand factor A1 domain. Science 2002;297: Wu Y, Suzuki-Inoue K, Satoh K, Asazuma N, Yatomi Y, Berndt MC, et al. Role of Fc receptor gamma-chain in platelet glycoprotein Ib-mediated signaling. Blood 2001;97: Mangin PH, Receveur N, Wurtz V, David T, Gachet C, Lanza F. Identification of five novel isoforms interacting with the GPIb-IX complex in platelets. J Thromb Haemost 2009; 7: Jain S, Zuka M, Liu J, Russell S, Dent J, Guerrero JA, et al. Platelet glycoprotein Ib alpha supports experimental lung metastasis. Proc Natl Acad Sci USA 2007;104: Yuan Y, Zhang W, Yan R, Liao Y, Zhao L, Ruan C, et al. Identification of a novel zeta binding site within the cytoplasmic domain of platelet glycoprotein Ibalpha that plays a key role in regulating the von Willebrand factor binding function of glycoprotein Ib-IX. Circ Res 2009;105: Mu FT, Cranmer SL, Andrews RK, Berndt MC. Functional association of phosphoinositide-3-kinase with platelet glycoprotein Ibalpha, the major ligand-binding subunit of the glycoprotein Ib-IX-V complex. J Thromb Haemost 2010;8: Andrews RK, Munday AD, Mitchell CA, Berndt MC. Interaction of calmodulin with the cytoplasmic domain of the platelet membrane glycoprotein Ib-IX-V complex. Blood 2001;98: Wu Y, Asazuma N, Satoh K, Yatomi Y, Takafuta T, Berndt MC, et al. Interaction between von Willebrand factor and glycoprotein Ib activates Src kinase in human platelets: role of phosphoinositide 3-kinase. Blood 2003;101: Kasirer-Friede A, Moran B, Nagrampa-Orje J, Swanson K, Ruggeri ZM, Schraven B, et al. ADAP is required for normal alphaiibbeta3 activation by VWF/GP Ib-IX-V and other agonists. Blood 2007;109: Schade AJ, Arya M, Gao S, Diz-Kucukkaya R, Anvari B, McIntire LV, et al. Cytoplasmic truncation of glycoprotein Ib

17 Thijs et al.: Normal platelet function S11 alpha weakens its interaction with von Willebrand factor and impairs cell adhesion. Biochemistry 2003;42: Gilbert JC, Feo-Fraulini T, Hutabarat RM, Horvath CJ, Merlino PG, Marsh HN, et al. First-in-human evaluation of anti von Willebrand factor therapeutic aptamer ARC1779 in healthy volunteers. Circulation 2007;116: Mayr FB, Knobl P, Jilma B, Siller-Matula JM, Wagner PG, Schaub RG, et al. The aptamer ARC1779 blocks von Willebrand factor-dependent platelet function in patients with thrombotic thrombocytopenic purpura ex vivo. Transfusion 2010;50: Bartunek J, Barbato E, Holz JB, Vercruysse K, Ulrichts H, Heyndrickx G. ALX-0081 a novel anti-thrombotic: results of a single-dose phase 1 study in healthy volunteers and further development in patients with stable angina undergoing PCI. Circulation 2008;118:S Vanhoorelbeke K, Ulrichts H, Van De Walle G, Fontayne A, Deckmyn H. Inhibition of platelet glycoprotein Ib and its antithrombotic potential. Curr Pharm Des 2007;13: Cauwenberghs N, Meiring M, Vauterin S, van Wyk V, Lamprecht S, Roodt JP, et al. Antithrombotic effect of platelet glycoprotein Ib-blocking monoclonal antibody Fab fragments in nonhuman primates. Arterioscler Thromb Vasc Biol 2000; 20: Nieswandt B, Bergmeier W, Schulte V, Rackebrandt K, Gessner JE, Zirngibl H. Expression and function of the mouse collagen receptor glycoprotein VI is strictly dependent on its association with the FcRgamma chain. J Biol Chem 2000; 275: Ezumi Y, Shindoh K, Tsuji M, Takayama H. Physical and functional association of the Src family kinases Fyn and Lyn with the collagen receptor glycoprotein VI-Fc receptor gamma chain complex on human platelets. J Exp Med 1998;188: Ichinohe T, Takayama H, Ezumi Y, Arai M, Yamamoto N, Takahashi H, et al. Collagen-stimulated activation of Syk but not c-src is severely compromised in human platelets lacking membrane glycoprotein VI. J Biol Chem 1997;272: Pasquet JM, Gross B, Quek L, Asazuma N, Zhang W, Sommers CL, et al. LAT is required for tyrosine phosphorylation of phospholipase cgamma2 and platelet activation by the collagen receptor GPVI. Mol Cell Biol 1999;19: Santoro SA. Identification of a 160,000 dalton platelet membrane protein that mediates the initial divalent cationdependent adhesion of platelets to collagen. Cell 1986;46: Staatz WD, Rajpara SM, Wayner EA, Carter WG, Santoro SA. The membrane glycoprotein Ia-IIa (VLA-2) complex mediates the Mgqq-dependent adhesion of platelets to collagen. J Cell Biol 1989;108: Van de Walle GR, Schoolmeester A, Iserbyt BF, Cosemans JM, Heemskerk JW, Hoylaerts MF, et al. Activation of alphaiibbeta3 is a sufficient but also an imperative prerequisite for activation of alpha2beta1 on platelets. Blood 2007; 109: Emsley J, Knight CG, Farndale RW, Barnes MJ, Liddington RC. Structural basis of collagen recognition by integrin alpha2beta1. Cell 2000;101: Knight CG, Morton LF, Onley DJ, Peachey AR, Messent AJ, Smethurst PA, et al. Identification in collagen type I of an integrin alpha2 beta1-binding site containing an essential GER sequence. J Biol Chem 1998;273: Raynal N, Hamaia SW, Siljander PR, Maddox B, Peachey AR, Fernandez R, et al. Use of synthetic peptides to locate novel integrin alpha2beta1-binding motifs in human collagen III. J Biol Chem 2006;281: Verkleij MW, IJsseldijk MJ, Heijnen-Snyder GJ, Huizinga EG, Morton LF, Knight CG, et al. Adhesive domains in the collagen III fragment alpha1(iii)cb4 that support alpha2beta1- and von Willebrand factor-mediated platelet adhesion under flow conditions. Thromb Haemost 1999;82: Xu Y, Gurusiddappa S, Rich RL, Owens RT, Keene DR, Mayne R, et al. Multiple binding sites in collagen type I for the integrins alpha1beta1 and alpha2beta1. J Biol Chem 2000;275: Iserbyt BF. Insights in platelet integrin alpha2beta1 affinity modulation. wdissertationx. Kortrijk: KU Leuven campus Kortrijk; Inoue O, Suzuki-Inoue K, Dean WL, Frampton J, Watson SP. Integrin alpha2beta1 mediates outside-in regulation of platelet spreading on collagen through activation of Src kinases and PLCgamma2. J Cell Biol 2003;160: Auger JM, Kuijpers MJ, Senis YA, Watson SP, Heemskerk JW. Adhesion of human and mouse platelets to collagen under shear: a unifying model. FASEB J 2005;19: Pugh N, Simpson AM, Smethurst PA, de Groot PG, Raynal N, Farndale RW. Synergism between platelet collagen receptors defined using receptor-specific collagen-mimetic peptide substrata in flowing blood. Blood 2010;115: De Meyer SF, Vanhoorelbeke K, Broos K, Salles II, Deckmyn H. Antiplatelet drugs. Br J Haematol 2008;142: Matsumoto Y, Takizawa H, Nakama K, Gong X, Yamada Y, Tandon NN, et al. Ex vivo evaluation of anti-gpvi antibody in cynomolgus monkeys: dissociation between anti-platelet aggregatory effect and bleeding time. Thromb Haemost 2006; 96: Ohlmann P, Hechler B, Ravanat C, Loyau S, Herrenschmidt N, Wanert F, et al. Ex vivo inhibition of thrombus formation by an anti-glycoprotein VI Fab fragment in non-human primates without modification of glycoprotein VI expression. J Thromb Haemost 2008;6: Grothusen C, Umbreen S, Konrad I, Stellos K, Schulz C, Schmidt B, et al. EXP3179 inhibits collagen-dependent platelet activation via glycoprotein receptor-vi independent of AT1-receptor antagonism: potential impact on atherothrombosis. Arterioscler Thromb Vasc Biol 2007;27: Ono K, Ueda H, Yoshizawa Y, Akazawa D, Tanimura R, Shimada I, et al. Structural basis for platelet antiaggregation by angiotensin II type 1 receptor antagonist losartan (DuP-753) via glycoprotein VI. J Med Chem 2010;53: Grosse J, Braun A, Varga-Szabo D, Beyersdorf N, Schneider B, Zeitlmann L, et al. An EF hand mutation in Stim1 causes premature platelet activation and bleeding in mice. J Clin Invest 2007;117: Bergmeier W, Oh-Hora M, McCarl CA, Roden RC, Bray PF, Feske S. R93W mutation in Orai1 causes impaired calcium influx in platelets. Blood 2009;113: Muller F, Mutch NJ, Schenk WA, Smith SA, Esterl L, Spronk HM, et al. Platelet polyphosphates are proinflammatory and procoagulant mediators in vivo. Cell 2009;139: Habib A, Vezza R, Creminon C, Maclouf J, FitzGerald GA. Rapid, agonist-dependent phosphorylation in vivo of human thromboxane receptor isoforms. Minimal involvement of protein kinase C. J Biol Chem 1997;272: Elzagallaai A, Rose SD, Trifaro JM. Platelet secretion induced by phorbol esters stimulation is mediated through phosphorylation of MARCKS: a MARCKS-derived peptide blocks

18 S12 Thijs et al.: Normal platelet function MARCKS phosphorylation and serotonin release without affecting pleckstrin phosphorylation. Blood 2000;95: Hoylaerts MF. Amplification loops: release reaction. In: Gresele P, Page CV, Fuster V, Vermeylen J, editors. Platelets in thrombotic and non-thrombotic disorders. Cambridge, UK: Cambridge University Press, 2002: Stefanini L, Roden RC, Bergmeier W. CalDAG-GEFI is at the nexus of calcium-dependent platelet activation. Blood 2009; 114: Lin LL, Wartmann M, Lin AY, Knopf JL, Seth A, Davis RJ. cpla2 is phosphorylated and activated by MAP kinase. Cell 1993;72: Moers A, Wettschureck N, Gruner S, Nieswandt B, Offermanns S. Unresponsiveness of platelets lacking both Galpha(q) and Galpha(13). Implications for collagen-induced platelet activation. J Biol Chem 2004;279: Offermanns S, Laugwitz KL, Spicher K, Schultz G. G-Proteins of the G(12) family are activated via thromboxane A(2) and thrombin receptors in human platelets. Proc Natl Acad Sci USA 1994;91: Loll PJ, Picot D, Garavito RM. The structural basis of aspirin activity inferred from the crystal structure of inactivated prostaglandin H2 synthase. Nat Struct Biol 1995;2: Lièvre M, Cucherat M. Aspirin in the secondary prevention of cardiovascular disease: an update of the APTC meta-analysis. Fundam Clin Pharmacol 2009;24: Hu H, Hoylaerts MF. The P2X1 ion channel in platelet function. Platelets 2010;21: Gould WR, Baxi SM, Schroeder R, Peng YW, Leadley RJ, Peterson JT, et al. Gas6 receptors Axl, Sky and Mer enhance platelet activation and regulate thrombotic responses. J Thromb Haemost 2005;3: Brass LF, Manning DR, Cichowski K, Abrams CS. Signaling through G proteins in platelets: to the integrins and beyond. Thromb Haemost 1997;78: Offermanns S. Activation of platelet function through G protein-coupled receptors. Circ Res 2006;99: Hechler B, Leon C, Vial C, Vigne P, Frelin C, Cazenave JP, et al. The P2Y1 receptor is necessary for adenosine 5 -diphosphate-induced platelet aggregation. Blood 1998;92: Ohlmann P, Laugwitz KL, Nurnberg B, Spicher K, Schultz G, Cazenave JP, et al. The human platelet ADP receptor activates Gi2 proteins. Biochem J 1995;312: Cattaneo M, Zighetti ML, Lombardi R, Martinez C, Lecchi A, Conley PB, et al. Molecular bases of defective signal transduction in the platelet P2Y12 receptor of a patient with congenital bleeding. Proc Natl Acad Sci USA 2003;100: Kahn ML, Zheng YW, Huang W, Bigornia V, Zeng D, Moff S, et al. A dual thrombin receptor system for platelet activation. Nature 1998;394: Vu TK, Hung DT, Wheaton VI, Coughlin SR. Molecular cloning of a functional thrombin receptor reveals a novel proteolytic mechanism of receptor activation. Cell 1991;64: Ishii K, Chen J, Ishii M, Koch WJ, Freedman NJ, Lefkowitz RJ, et al. Inhibition of thrombin receptor signaling by a G- protein coupled receptor kinase. Functional specificity among G-protein coupled receptor kinases. J Biol Chem 1994;269: De Candia E, Hall SW, Rutella S, Landolfi R, Andrews RK, De Cristofaro R. Binding of thrombin to glycoprotein Ib accelerates the hydrolysis of Par-1 on intact platelets. J Biol Chem 2001;276: Siller-Matula JM, Krumphuber J, Jilma B. Pharmacokinetic, pharmacodynamic and clinical profile of novel antiplatelet drugs targeting vascular diseases. Br J Pharmacol 2010;159: Woulfe D, Yang J, Prevost N, O Brien PJ, Brass LF. Signal transduction during the initiation, extension and perpetuation of platelet plug formation. In: Michelson AD, editor. Platelets. London: Elsevier Science Academic Press, 2002: Jhon DY, Lee HH, Park D, Lee CW, Lee KH, Yoo OJ, et al. Cloning, sequencing, purification, and Gq-dependent activation of phospholipase C-beta 3. J Biol Chem 1993;268: Park D, Jhon DY, Kriz R, Knopf J, Rhee SG. Cloning, sequencing, expression, and Gq-independent activation of phospholipase C-beta 2. J Biol Chem 1992;267: Klages B, Brandt U, Simon MI, Schultz G, Offermanns S. Activation of G12/G13 results in shape change and Rho/Rhokinase-mediated myosin light chain phosphorylation in mouse platelets. J Cell Biol 1999;144: Yang J, Wu J, Jiang H, Mortensen R, Austin S, Manning DR, et al. Signaling through Gi family members in platelets. Redundancy and specificity in the regulation of adenylyl cyclase and other effectors. J Biol Chem 2002;277: Cheng Y, Austin SC, Rocca B, Koller BH, Coffman TM, Grosser T, et al. Role of prostacyclin in the cardiovascular response to thromboxane A2. Science 2002;296: Bevers EM, Comfurius P, Vanrijn JL, Hemker HC, Zwaal RF. Generation of prothrombin-converting activity and the exposure of phosphatidylserine at the outer surface of platelets. Eur J Biochem 1982;122: Hato T, Pampori N, Shattil SJ. Complementary roles for receptor clustering and conformational change in the adhesive and signaling functions of integrin alphaiib beta3. J Cell Biol 1998;141: Arnaout MA, Goodman SL, Xiong JP. Structure and mechanics of integrin-based cell adhesion. Curr Opin Cell Biol 2007; 19: Calderwood DA, Fujioka Y, de Pereda JM, Garcia-Alvarez B, Nakamoto T, Margolis B, et al. Integrin beta cytoplasmic domain interactions with phosphotyrosine-binding domains: a structural prototype for diversity in integrin signaling. Proc Natl Acad Sci USA 2003;100: Calderwood DA, Yan BX, de Pereda JM, Alvarez BG, Fujioka Y, Liddington RC, et al. The phosphotyrosine binding-like domain of talin activates integrins. J Biol Chem 2002;277: Ma YQ, Qin J, Wu CY, Plow EF. Kindlin-2 (Mig-2): a coactivator of beta(3) integrins. J Cell Biol 2008;181: Banno A, Ginsberg MH. Integrin activation. Biochem Soc Trans 2008;36: Di Paolo G, Pellegrini L, Letinic K, Cestra G, Zoncu R, Voronov S, et al. Recruitment and regulation of phosphatidylinositol phosphate kinase type 1 gamma by the FERM domain of talin. Nature 2002;420: Lee HS, Lim CJ, Puzon-McLaughlin W, Shattil SJ, Ginsberg MH. RIAM activates integrins by linking talin to ras GTPase membrane-targeting sequences. J Biol Chem 2009;284: Harburger DS, Bouaouina M, Calderwood DA. Kindlin-1 and- 2 directly bind the C-terminal region of beta integrin cytoplasmic tails and exert integrin-specific activation effects. J Biol Chem 2009;284:

19 Thijs et al.: Normal platelet function S Moser M, Nieswandt B, Ussar S, Pozgajova M, Fassler R. Kindlin-3 is essential for integrin activation and platelet aggregation. Nat Med 2008;14: Nieswandt B, Moser M, Pleines I, Varga-Szabo D, Monkley S, Critchley D, et al. Loss of talin1 in platelets abrogates integrin activation, platelet aggregation, and thrombus formation in vitro and in vivo. J Exp Med 2007;204: Moser M, Legate KR, Zent R, Fassler R. The tail of integrins, talin, and kindlins. Science 2009;324: Yang Z, Kollman JM, Pandi L, Doolittle RF. Crystal structure of native chicken fibrinogen at 2.7 A resolution. Biochemistry 2001;40: Bottcher RT, Lange A, Fassler R. How ILK and kindlins cooperate to orchestrate integrin signaling. Curr Opin Cell Biol 2009;21: Lai-Cheong JE, Ussar S, Arita K, Hart IR, McGrath JA. Colocalization of kindlin-1 kindlin-2, and migfilin at keratinocyte focal adhesion and relevance to the pathophysiology of kindler syndrome. J Invest Dermatol 2008;128: Tu YZ, Wu S, Shi XH, Chen K, Wu CY. Migfilin and Mig- 2 link focal adhesions to filamin and the actin cytoskeleton and function in cell shape modulation. Cell 2003;113: Prevost N, Woulfe DS, Jiang H, Stalker TJ, Marchese P, Ruggeri ZM, et al. Eph kinases and ephrins support thrombus growth and stability by regulating integrin outside-in signaling in platelets. Proc Natl Acad Sci USA 2005;102: Saltzman AJ, Mehran R, Hooper WC, Moses JW, Weisz G, Collins MB, et al. The relative effects of abciximab and tirofiban on platelet inhibition and C-reactive protein during coronary intervention. J Invasive Cardiol 2010;22: Brener SJ. The benefits of platelet glycoprotein IIb/IIIa receptor inhibition during primary percutaneous coronary intervention for ST-segment elevation myocardial infarction: drugspecific or class effect? wcommentx J Am Coll Cardiol 2009;53: Comment on: J Am Coll Cardiol 2009;53: George JN, Aster RH. Drug-induced thrombocytopenia: pathogenesis, evaluation, and management. Hematology Am Soc Hematol Educ Program 2009; Arehart E, Stitham J, Asselbergs FW, Douville K, MacKenzie T, Fetalvero KM, et al. Acceleration of cardiovascular disease by a dysfunctional prostacyclin receptor mutation: potential implications for cyclooxygenase-2 inhibition. Circ Res 2008; 102: Gambaryan S, Kobsar A, Rukoyatkina N, Herterich S, Geiger J, Smolenski A, et al. Thrombin and collagen induce a feedback inhibitory signaling pathway in platelets involving dissociation of the catalytic subunit of PKA from an NF-{kappa}B-I{kappa}B complex. J Biol Chem 2010;285: Aszodi A, Pfeifer A, Ahmad M, Glauner M, Zhou XH, Ny L, et al. The vasodilator-stimulated phosphoprotein (VASP) is involved in cgmp- and camp-mediated inhibition of agonistinduced platelet aggregation, but is dispensable for smooth muscle function. EMBO J 1999;18: Pula G, Schuh K, Nakayama K, Nakayama KI, Walter U, Poole AW. PKCdelta regulates collagen-induced platelet aggregation through inhibition of VASP-mediated filopodia formation. Blood 2006;108: Bauer M, Retzer M, Wilde JI, Maschberger P, Essler M, Aepfelbacher M, et al. Dichotomous regulation of myosin phosphorylation and shape change by Rho-kinase and calcium in intact human platelets. Blood 1999;94: Hathaway DR, Eaton CR, Adelstein RS. Regulation of human platelet myosin light chain kinase by the catalytic subunit of cyclic AMP-dependent protein kinase. Nature 1981;291: Walter U, Eigenthaler M, Geiger J, Reinhard M. Role of cyclic nucleotide-dependent protein kinases and their common substrate VASP in the regulation of human platelets. Adv Exp Med Biol 1993;344: Naseem KM. enos, inos or no NOS, that is the question! wcommentx J Thromb Haemost 2008;6: Comment on: J Thromb Haemost 2008;8: Schlossmann J, Ammendola A, Ashman K, Zong X, Huber A, Neubauer G, et al. Regulation of intracellular calcium by a signalling complex of IRAG, IP3 receptor and cgmp kinase Ibeta. Nature 2000;404: Trepakova ES, Cohen RA, Bolotina VM. Nitric oxide inhibits capacitative cation influx in human platelets by promoting sarcoplasmic/endoplasmic reticulum Ca 2q -ATPase-dependent refilling of Ca 2q stores. Circ Res 1999;84: Wang GR, Zhu Y, Halushka PV, Lincoln TM, Mendelsohn ME. Mechanism of platelet inhibition by nitric oxide: in vivo phosphorylation of thromboxane receptor by cyclic GMPdependent protein kinase. Proc Natl Acad Sci USA 1998;95: Maurice DH, Haslam RJ. Molecular basis of the synergistic inhibition of platelet function by nitrovasodilators and activators of adenylate cyclase: inhibition of cyclic AMP breakdown by cyclic GMP. Mol Pharmacol 1990;37: Pigazzi A, Heydrick S, Folli F, Benoit S, Michelson A, Loscalzo J. Nitric oxide inhibits thrombin receptor-activating peptide-induced phosphoinositide 3-kinase activity in human platelets. J Biol Chem 1999;274: Massberg S, Gruner S, Konrad I, Garcia Arguinzonis MI, Eigenthaler M, Hemler K, et al. Enhanced in vivo platelet adhesion in vasodilator-stimulated phosphoprotein (VASP)- deficient mice. Blood 2004;103: Rajendran S, Chirkov YY. Platelet hyperaggregability: impaired responsiveness to nitric oxide ( platelet NO resistance ) as a therapeutic target. Cardiovasc Drugs Ther 2008; 22: Lippi G, Favaloro EJ. The missing link between genotype, phenotype and clinics. Biochemia Medica 2009;19: Zuern CS, Schwab M, Gawaz M, Geisler T. Platelet pharmacogenomics. J Thromb Haemost 2010;8: wepub 2010 Feb 1x.

20

21 Clin Chem Lab Med 2010;48(Suppl 1):S15 S by Walter de Gruyter Berlin New York. DOI /CCLM Review Hypercoagulable state, pathophysiology, classification and epidemiology Zrinka Alfirević 1, * and Igor Alfirević 2 1 Department of Internal Medicine, Medical School, University Hospital Sestre Milosrdnice, Zagreb, Croatia 2 Special Hospital for Cardiovascular Surgery and Cardiology Magdalena Krapinske Toplice, Croatia Abstract Hypercoagulable state is not a uniform disease. It is a complex condition with an abnormal propensity for thrombosis that may or may not lead to thrombosis, depending on complex gene-gene and gene-environment interactions. The prevalence of the hypercoagulable state depends on the ethnicity and clinical history of the population being studied. The consequences of a hypercoagulable state due to thrombosis of veins and arteries are the most important cause of sickness and death in developed countries at present. Primary hypercoagulable state is an inherited condition caused by the reduced level of natural anticoagulants due to a qualitative defect or quantitative deficiency of an antithrombotic protein, or increased concentrations or function of coagulation factors. Most of the inherited abnormalities recognized to date have little or no effect on arterial thrombosis and are associated primarily with venous thromboembolism. Arterial thrombosis usually develops as a complication of atherosclerosis and patients usually have more than one traditional risk factor. Secondary hypercoagulable states generally occur as a result of a large number of transient or permanent acquired conditions that increase the tendency for formation of blood clots. New epidemiological data and clinical trials suggest that many acquired risk factors in the pathophysiology of arterial and venous thrombosis overlap and coexist for both disorders. Clin Chem Lab Med 2010;48:S Keywords: acquired conditions; arterial and venous thrombosis; hypercoagulable state; inherited thrombophilia. Introduction The hypercoagulable states are a group of inherited or acquired conditions that cause a pathological thrombotic ten- *Corresponding author: Zrinka Alfirević, Department of Internal Medicine, Medical School University Hospital Sestre Milosrdnice, Vinogradska 29, 10,000 Zagreb, Croatia Phone: q , zrinka.alfirevic@gmail.com Received June 16, 2010; accepted October 16, 2010; previously published online November 16, 2010 dency due to an abnormality in the blood itself or changes in vasculature. There are no specific signs or symptoms associated with hypercoagulable states. The most common clinical manifestation is venous thromboembolic disease, although hypercoagulable states increase the risk for both arterial and venous thrombosis. Some conditions are associated to a greater extent with venous, and others with arterial thrombosis, but there are also conditions that are associated with the development of both arterial and venous thrombosis (1). The term thrombosis refers to the formation, from constituents in the blood, of a mass within the venous or arterial vasculature of a living being (2). The presence of hypercoagulability is not associated uniquely with the development of thrombosis, and there is no data to suggest that hypercoagulable states reduce survival in patients who carry an inherited predisposition to hypercoagulability. A person with a well-defined hypercoagulable state does not necessarily develop thrombosis, and not all people with thrombosis have an identifiable hypercoagulable state. Many of these patients with inherited defects are asymptomatic until they experience some acquired risk factor. There is no specific therapy and we do not cure the cause, but the consequences. Antiplatelets, fibrinolytics and anticoagulants are utilized to treat arterial thromboembolism (ATE), and anticoagulants and fibrinolytics may be utilized to treat venous thromboembolism (VTE). Patients are considered to have hypercoagulable states if they have laboratory abnormalities or clinical conditions that are associated with an increased risk of thrombosis (prethrombotic states), or if they have recurrent thrombosis without recognizable predisposing factors (i.e., they are thrombosis-prone). Among the factors that may indicate a primary hypercoagulable state are a family history of thrombosis, recurrent thrombosis without apparent precipitating factors, thrombosis at unusual anatomic sites, thrombosis at an early age, and resistance to conventional antithrombotic therapy. In everyday practice, accurate determination of causes that have led to a hypercoagulable state is useful in deciding on the length of therapy. Testing should also be directed at identifying any underlying systemic disorder since treatment may improve the thrombotic tendency (1, 3). Today, we know that the development and location of thrombosis depends on a wide range of diseases, clinical circumstances or molecular variants associated with an increased risk of thrombosis. However, the cause remains unknown in many patients. Thrombus formation depends on complex interactions between heritage and the gene-environment relationship (4). 2010/356

22 S16 Alfirević and Alfirević: Conditions associated with hypercoagulable state Arterial thrombosis usually develops as a complication of atherosclerosis resulting from hypertension, hyperlipidaemia, diabetes mellitus, metabolic syndrome and homocystinuria. These patients have hypercholesterolaemia and other lipoprotein abnormalities, high fibrinogen concentrations and increased von Willebrand factor (vwf), tissue plasminogen activator antigen and factor VIII activity (5). Coronary arterial thrombosis and cerebrovascular thrombotic occlusion are the leading causes of morbidity and mortality in developed countries. Myocardial infarction and thrombotic stroke cause more than one third of all deaths in developed countries (6). In contrast, venous blood stasis initiates local activation of procoagulant events to form venous thrombus by accumulation of fibrin and trapped erythrocytes. Hypercoagulability due to inherited genetic abnormalities plays a crucial role in the development of venous thrombosis. Thrombophilia is generally associated with venous thrombosis and may be detected in a significant number of patients who present with their first episode of venous thromboembolism. The term thrombophilia, as opposed to haemophilia, refers to inherited or acquired conditions in the coagulation system, with thrombosis as their primary manifestation and is often not evident without specific laboratory testing. These patients are at higher risk for thrombosis as compared with the normal population, and the risk increases according to the number of risk factors (6). Venous thromboembolic disease is the third leading cause of cardiovascular death in developed countries. It causes 50, ,000 deaths in the US annually (7). The disease can manifest itself through thrombotic obstruction of any venous system and usually occurs as deep vein thrombosis of the leg. An unusual venous location occurs in fewer than 5% of all cases of thrombosis, and is frequently associated with some genetic defect. Foetal complications in the form of massive pulmonary embolism occurs in 20% of cases (8, 9). From the first review article registered in 1964 in the Medline database, the view on hypercoagulable states has been updated continuously (10). Interest in these disorders has increased in the last 50 years due to the discovery of several genetic factors that contribute to the incidence of thrombosis. The initial report was published in 1965 by Egeberg who described a Norwegian family with an identified hereditary tendency to thrombosis, caused by antithrombin deficiency (11). From 1965 to 1993, a heritable cause of thrombosis was detected in 5% 15% of patients with thromboembolic disease. Today, inherited thrombophilia can be found in over 60% of cases following the first clinical episode of VTE (12). This article will focus on the pathophysiological mechanisms of arterial and venous thrombosis, and underlying inherited and acquired conditions in epidemiological situations. Haemostasis The human haemostatic system maintains blood fluidity on the basis of a well-controlled interaction between platelets, the vascular endothelium and protein (procoagulant, natural anticoagulant, fibrinolytic and antifibrinolytic) components. Haemostatic thrombosis in a normal healthy organism is a self-limited, well-controlled condition that prevents leakage of blood from the blood vessel. It is subdivided into four well-known conditions: vascular constriction, primary and secondary haemostasis, and dissolution of the clot. Immediately after vascular injury, vascular vasoconstriction limits the flow of blood to the area of injury. At the same time damage to the blood vessel walls exposes underlying subendothelium proteins (vwf) to blood proteins, initiating primary haemostasis (13). The aim of primary haemostasis is formation of a platelet plug. Circulating platelets adhere to damaged endothelium with the help of von Willebrand factor, by binding to collagen where they enlarge and flatten. Adhesion, activation and aggregation of platelets is a process mediated by several factors secreted by surrounding cells and by platelets themselves. These mediators activate other platelets to link by fibrinogen through the surface glycoprotein IIb/IIIa receptors to form the platelet plug, and further activate vasoconstriction and the clotting cascade to form an organized clot (14, 15). The aim of secondary haemostasis is to continue and complete the process of haemostasis through a series of wellknown reactions known as the coagulation cascade. Secondary haemostasis can take place through two common pathways initiated by distinct mechanisms which lead to formation of a thrombus: the tissue factor pathway (formerly known as the extrinsic pathway) and the contact activation pathway (formerly known as the intrinsic pathway). In both pathways, inactive enzyme precursor of a serine protease and its glycoprotein co-factor are activated to become active components that then catalyse the next reaction in the cascade. This results ultimately in cross-linked fibrin which stabilizes a platelet plug and forms a durable thrombus. The intrinsic pathway has low significance under normal physiological conditions; its activation begins when negatively charged collagen inside the damaged endothelium, or negatively charged phospholipids on the surface of activated platelets, activate formation of the primary complex on collagen by high-molecular-weight kininogen (HMWK), prekallikrein, and FXII (Hageman factor). The main pathway of in vivo coagulation takes place through tissue factor. Tissue factor is the only protein of coagulation which does not circulate in the blood. As non-functional precursors, it is fully functional only in cases of cell damage when it is expressed on the surface of endothelial cells, platelets or leukocytes. It is present on almost all extra-vascular cells, with the exception of joint coatings. Also, it may be expressed during physiological processes, such as normal parturition. The coagulation cascade occurs during thrombus formation in response to tissue injury. Activation begins with the release of tissue factor from injured cells where activated factor VII, expressed on tissue-factor-bearing cells, forms an activated complex (TF-FVIIa). This complex further activates factor X and leads to formation of a small amount of thrombin. Factor VIIa can also form a tenase complex through activation of factor XI, which then further initiates the activation of factor IX. Factor IX, together with activated factor VIII,

23 Alfirević and Alfirević: Conditions associated with hypercoagulable state S17 activates factor X. Both pathways convert to activation of thrombin, the key enzyme of haemostasis. The common point is factor X. Factor Xa activates prothrombin (factor II) which is converted to thrombin (factor IIa). The activation of thrombin occurs on the surface of activated platelets and requires formation of a prothrombinase complex. This complex is composed of platelet phospholipids, phosphatidylinositol and phosphatidylserine, Ca 2q, factors Va and Xa, and prothrombin. Thrombin converts fibrinogen to fibrin, and also activates factors VIII and V, and their inhibitor, protein C (in the presence of thrombomodulin). Factor XIII acts on fibrin polymers to form activated monomers which form bridges between aggregated and adjacent stimulated platelets and the stabilized platelet plug (16 18). The procoagulation system is very powerful, and potentially dangerous. Therefore, the production of thrombin, its generation from prothrombin, and its activity in plasma are carefully regulated by thromboregulatory anticoagulants at each step of the cascade. Natural anticoagulants, such as protein C, protein S, antithrombin III, the tissue factor pathway inhibitor (TFPI) and protein Z, inhibit activated enzymes and associated cofactors to prevent pathological thrombosis, limiting the clotting process to the actual site of injury. However, despite all the mechanisms of control, if thrombus formation occurs, conter regulatory balancing mechanisms from the fibrinolytic system attempt to reduce and eliminate the thrombi. Plasmin is the most important enzyme of physiological fibrinolysis and limits thrombus size and dissolves a thrombus by degrading a cross-linked fibrin once the vascular injury has been repaired (19). Pathophysiology Abnormalities in blood flow due to turbulence (aneurysm, hypertension), blood stasis (immobilization, tumour compression) or hyperviscosity (myeloproliferative disorder, thrombocytosis) may contribute to thrombogenesis due to local hypoxia which stimulates endothelial cells to release an activator of factor X (20). Disturbances of the vessel wall (vascular endothelial damage) may occur as a result of direct physical trauma during surgical procedures, or as a result of local exposure to proinflammatory or procoagulant mediators (21). Normal endothelium is physiologically non-thrombogenic and responsible for normal blood fluidity. Endothelial cells defend themselves and control thromboregulation through continuous communication with platelets and with several of their own mechanisms, such as the thrombomodulin protein C system, the tissue factor pathway inhibitor system, the plasmin generating system and antithrombin system. Endothelial prostacyclins, ADP and nitric oxide inhibit platelet aggregation and support vascular relaxation, endogenous heparin, and block thrombin production. At the same time, prostacyclin released locally from platelets inhibit platelet adhesion and aggregation. In situations where damage occurs, endothelial cells release endothelin-1 and platelet activating factor, increase production of vwf, PAI-1 and factor V, and also produce one of the strongest drivers of haemostasis, a cellular membrane protein known as tissue factor. Coagulation can also be activated by factor X, activated platelets, monocytes or factor XII (22). Hypercoagulability is the propensity to develop thrombosis due to an abnormality of any haemostatic cellular and protein component caused by a hyperactive procoagulant or hypoactive anticoagulant and/or defect in fibrinolysis. Hypercoagulability is due to disruption of the highly regulated coagulation mechanism and occur as a result of hereditary defects in one or more of clotting factors, or some situational or acquired condition. The level of lifelong, baseline hypercoagulability in any individual may be determined by the type and number of defects that are inherited. The inheritance of more than one genetic polymorphism increases the risk (23). Environmental risk factors, such as pregnancy, use of oral contraceptives and heparin, induce thrombocytopenia and represent transient circumstances, whereas increased risk lasts as long as the disorder that has caused it to occur. Acquired risk factors are, for the most part, the result of a non-reversible medical disorder, such as cancer, the nephrotic syndrome, myeloproliferative syndrome and paroxysmal nocturnal haemoglobinuria (24). Arterial thrombosis Arterial thrombus formation usually begins with endothelial injury (atherosclerotic plaque rupture, aneurysm formation, vessel dissection, hyperhomocysteinaemia) in a situation where there is high velocity blood flow and increased platelet activity. Hyperviscosity due to myeloproliferative syndromes, cryoglobulinemia and plasma cell dyscrasias, may also precipitate arterial thrombosis. The core of an atherosclerotic plaque is rich in inflammatory cells, lipids, cholesterol crystals, and tissue factors generated by activated macrophages. In cases of plaque ulceration, the arterial endothelium becomes damaged and highly thrombogenic lipid components are exposed to the bloodstream. These thrombogenic lipid components activate adherent neutrophils and platelets to cause release of inflammatory and procoagulant mediators that further amplify thrombosis (5). In this case, platelet granules release adenosine diphosphate, adenosine triphosphate, calcium and serotonin and induce additional platelets in the surrounding milieu to activate and aggregate. Platelet activation by ADP leads to a conformational change in platelet glycoprotein IIb/IIIa receptors that induces them to bind to fibrinogen (25). Stimulated platelets bind related adhesive proteins of vwf, fibronectin, and vitronectin. Next, the GpIIb/IIIa-bound protein cross-links these platelets into aggregates that bind to the layer of adherent platelets, forming a haemostatic plug or an occlusive thrombus (26). An arterial (white) thrombus is formed primarily from platelets. Its formation resembles a normal haemostatic plaque except that it is located on the external surface of a ruptured atherosclerotic plaque. Thrombus growth and fibrin deposition may produce a red thrombus and totally occlude the flow of

24 S18 Alfirević and Alfirević: Conditions associated with hypercoagulable state blood. This can result in transition from stable or subclinical atherosclerotic disease to acute myocardial infarction, stroke, or peripheral arterial occlusion (27). Most arterial thromboses develop on the basis of endothelial injuries in association with traditional cardiovascular risk factors which themself are potent procoagulant factors (28). Genetic abnormalities in platelet function due to polymorphisms in platelet glycoprotein IIIa polypeptide, high fibrinogen concentrations, increased plasma vwf concentrations, and increased concentrations of factors VII, VIII, IX, XI, XII are also responsible for hypercoagulability state in the pathophysiology of arterial thrombosis. Any qualitative or quantitative changes in platelet count or function due, for example, to a polymorphism in the b-subunit of G protein increases prothrombotic risk (29). A strong association between the (gp)iiia, (Pl A2 ), (gp)ia (8807T allele), and (gp)iba polymorphism of the glycoprotein IIIa gene due to altered adhesion, spreading and actin cytoskeletal rearrangement and acute coronary thrombosis has been described in patients who had coronary events before 60 years of age. The odds ratio for coronary thrombosis is six-fold higher in patients younger than 60 years who have a glycoprotein IIb/ IIIa allele (30). Sticky platelet syndrome is an example of a congenital, autosomal dominant disorder of platelets, characterized by hyperaggregable platelets in response to ADP and epinephrine, and associated with arterial and venous thromboembolic event (31). An increased fibrinogen concentration is a strong and independent predictor of cardiovascular risk, both in healthy individuals and in those with manifest coronary artery disease (32 34). Many studies have found a connection between increased plasma factor VII activity and arterial thrombosis (35). Other disorders of the fibrinolytic system due to defective synthesis of tissue plasminogen activator or deficiency or a functional defect in the plasminogen molecule can also be a possible cause of a hypercoagulable state in the pathogenesis of arterial thrombosis. The plasminogen activator inhibitor-1 (PAI-1) promoter 4G/5G insertion/deletion polymorphism has been associated with 25% higher plasma PAI-1 activity. The prevalence of the 4G allele is significantly higher in patients with myocardial infarction before the age of 45 years than in population-based controls, (allele frequencies of 0.63 vs. 0.53). Both alleles bind a transcriptional activator. Increased gene transcription is associated with four guanine bases (the 4G allele), and results in the binding of a transcriptional activator alone, whereas the 5G allele also binds a repressor protein that decreases the binding of the activator (36). Venous thrombosis In contrast, venous blood stasis is known to initiate the formation of a venous thrombus by the local activation of procoagulant zymogens. A red venous thrombus is composed of large amounts of fibrin that contains entrapped erythrocytes, and very often occurs in intact endothelium. Exceptions include direct venous trauma, extrinsic venous compression, and biochemical vascular endothelial cell injury resulting from the toxic effect of many prothrombotic mediators (37). Venous stasis due to immobilization, hospitalization, limb paralysis, pregnancy, varicose vein, or venous insufficiency increases venous retrograde pressure and potentiates local hypoxia. Damaged endothelial cells release prothrombotic and proinflammatory mediators. The process begins with tissue factor and leads to the sequential activation of coagulation factor X (FXa), cleavage of prothrombin by FXa to liberate thrombin, and cleavage of fibrinogen by thrombin to produce fibrin monomers. Fibrin monomers spontaneously polymerise into fibrin strands that are then irreversibly crosslinked by thrombin-activated factor XIII to produce an insoluble clot. Thrombosis can occur either when the production of FXa and thrombin is enhanced or when the activity of control proteins is diminished. Endothelial cell injury increases the surface expression of cell adhesion molecules, promoting leukocyte and platelet adhesion and activation. This initiates and amplifies inflammation and thrombosis (38, 39). Hypercoagulability, due to inherited genetic abnormalities, plays a crucial role in the development of venous thrombosis (40). These conditions are characterized by disruption in the normally highly regulated coagulation mechanism, resulting in increased thrombin generation and an increased risk of clinical thrombosis. In normal conditions, clotting factors are deceptive and wait for a signal for activation (41, 21). Activated protein C (APC) resistance due to a single point mutation in the factor V gene is the most common relatively mild inherited disorder that causes hypercoagulability. It is responsible for 20% 60% of cases of thrombosis in Caucasians. The risk in carriers of one polymorphic allele of Factor V Leiden (FV Leiden) is 4 8/ 1000, and for homozygotes the risk is 80/1000 per year (42). APC resistance is a laboratory clotting parameter, but it is also a clinical name for familial thrombotic disorder, a condition where the patient s plasma does not produce an appropriate anticoagulant response to APC. In 90% of cases, patients have the factor V allele that is resistant to the proteolytic effect of protein C. FV Leiden is a factor V variant which is APC inactivated 10 times more slowly (43). The second most common cause of inherited thrombophilia is a mutation in prothrombin. People with the prothrombin G20210A mutation have higher amounts of prothrombin than normal due to a decreased rate of degradation of mrna, with a three to seven-fold increased risk of thrombosis in heterozygotes. Compound heterozygosity with FV Leiden increases the risk of VTE 20-fold (44, 45). A deficiency in natural anticoagulants is a strong, but rare, prothrombotic risk factor. Severe AT deficiency may increase the risk of VTE up to 50-fold (46). The patients heterozygous for protein C deficiency have a 10-fold increased risk for thrombosis, develop thrombosis between the age of 15 and 30 years, and have 60% less than normal antigenic protein C (47, 48). Increased factor VIII (FVIII) concentrations have also been associated with increased prothrombotic risk (49). Epidemiological data have shown that high concentrations of factors IX and XI are also associated with an increased risk of venous thrombosis, but an adequate association has not been well-established (50, 51).

25 Alfirević and Alfirević: Conditions associated with hypercoagulable state S19 Deficiency of any other fibrinolytic com-ponent may be a potential cause of venous thrombosis, but arterial events may also occur. There are five forms of dysfibrinolysis: congenital plasminogen deficiency, congenital deficiency of tissue plasminogen activator or congenital increases in plasminogen activator inhibitor, congenital dysfibrinogenaemia and factor XII deficiency. Plasminogen deficiency is an autosomal dominant disorder caused by mutations of the PLG gene; thrombotic events occur when plasminogen values are -40% of normal (52). Deficiency of heparin cofactor II, deficiency of tissue factor pathway inhibitor, and thrombomodulin mutations are extremely rare, but are potential prothrombotic risk factors. Classification According to the mechanism of action and time of occurrence, the hypercoagulable state can be divided into an inherited or acquired condition, primary or secondary, severe or rare and permanent or transient. Some states are associated more with arterial thrombosis, while others are associated with venous thrombosis (Table 1) (1, 53). Hypercoagulable states are usually divided into primary and secondary states. Until 1993, the primary hypercoagulable states included antithrombin III, protein C and protein S deficiencies, dysfibrinogenemias, plasminogen deficiency, and decreased plasminogen activator activity I (53). Since then, the number of specific genetic polymorphisms that play a role in the pathogenesis of arterial and venous thrombosis have been rising. It was recognized that most, if not all, patients with venous thromboembolism have a genetic basis for the disorder, i.e., thrombophilia (54). Thrombophilia is an inherited or acquired clinical phenotype manifesting in selected individuals as a greater risk for the development of recurrent thrombosis at a younger age compared with the general population, with considerable differences in the magnitude of risk among individuals in the same family and with the same thrombophilic gene defect. Hereditary thrombophilia is a condition with thrombosis as its primary manifestation, and is often not evident without specific laboratory testing to reveal its mechanism of action. After 1993, the term thrombophilia has often been identified as the primary hypercoagulable state (55, 56). There are 21,000 published articles in the Medline electronic data base associated with the hypercoagulable state. Most of the articles are focused on a single association of each factor with venous thrombosis, and/or the role of thrombophilia in the pathogenesis of venous thrombosis. Only a few of the articles generally address hypercoagulable states (57, 58) and mentioned con- Table 1 Genetic disorders associated with hypercoagulable states. Hypercoagulable states Disorders Arterial Both arterial and venous Venous Inherited Increased fibrinogen concentrations PAI-1 excess Definite (a, b chain polymorphism) Dysfibrinogenaemia AT Platelet glycoprotein polymorphism Homocystinuria Protein C GPIIIa Leu 33Pro Protein S GP1BA 5T/C Factor V G1691A (APC) GP&1325T/C PT G20210A Thrombomodulin deficiency Elevated factor VIII Lipoprotein(a) Possible Factor VII excess Factors XI, IX, TAFI, von Willebrand, heparin cofactor II Mixed Hyperhomocysteinaemia Acquired Hypertension Antiphospholipid antibody Previous thrombosis Smoking Cancer Immobilization Diabetes melitus Age Major surgery Hypercholesterolaemia Obesity Orthopedic surgery Metabolic syndrome Pregnancy and puerperium HITTS Nephrotic syndrome Malignancy Oral contraceptives Hormone replacement therapy Infection Myeloproliferative disorders (essential thrombocytosis polycythaemia rubra vera, chronic myeloid leukaemia, myelofibrosis) Paroxysmal nocturnal hemoglobinuria

26 S20 Alfirević and Alfirević: Conditions associated with hypercoagulable state ditions that are associated with both arterial and venous thrombosis. Thus, it may be concluded that classification differs slightly from author to author. Primary hypercoagulable states All authors agree that the primary hypercoagulable states include inherited conditions caused by decreased amounts of natural anticoagulants due to a qualitative defect or quantitative deficiency of an antithrombotic protein, or by increased amounts or function of coagulation factors. Most of the inherited abnormalities recognized to date have little or no effect on arterial thrombosis, and are associated primarily with venous thromboembolism (55). Reduced concentrations of the inhibitors of the coagulation cascade involve a rare but strong risk of inherited mutations (deficiency of antithrombin III, protein C and protein S) commonly found in Southeast Asia, and are rare in the black population. Deficiency is divided into subgroups according to antigen and activity levels. Most patients will have an episode of thrombosis by 60 years of age. Homozygous deficiencies are rare and can result in a severe phenotype with neonatal purpura fulminans (deficiency of PC and PS), or are incompatible with normal life (antithrombin deficiency) (48, 55). At present, the prothrombin G20210A mutations or APC resistance due to FV Leiden are the most prevalent causes of thrombosis in Caucasians (59). The risk is mild and increases with age and in proportion to the number of risk factors. Most of these individuals will not have had an episode of venous thrombosis by the age of 60 years, or the disorder will be expressed by exposure to another risk factor (43, 44). Not as a result of an acute phase reaction, increased factor VIII concentrations are also a risk factor for venous thrombosis, particularly in the black population (60, 61). Some studies suggest that a weak association exists between FV Leiden and prothrombin G20210A arterial thrombosis (62). A number of less frequent hereditary abnormalities (increased concentrations of factors IX and XI, VII, von Willebrand, TAFI, deficiency of heparin cofactor II) have also been associated with an increased tendency for venous thrombosis. However, such associations have not been wellestablished, even though the relevance varies from author to author. For example, Schafer mentioned, among other factors, the increased levels of factors VII, XI, IX, and von Willebrand in the primary hypercoagulable state. Nachman, in the primary condition, included MTHFR mutation that is rarely mentioned by other authors, but dysfibrinogenemia, mentioned by many other authors, is not listed (57). The pathophysiological mechanism of the primary hypercoagulable state is a result of impaired regulation of endothelial function (AT III, PC, PS) or alterations in the kinetics of coagulation zymogens (FV Leiden, prothrombin G20210A) that are associated with surface expression of tissue factor that lead to venous thrombosis. Hereditary thrombophilia factors are clearly associated with venous thrombosis; the exception is dysfibrinogenaemia which also plays a role in the development of arterial thrombosis (57). When discussing the primary hypercoagulable state, we should also mention the rare genetic disorders that play a role in the pathophysiology of arterial thrombosis (platelet glycoprotein polymorphism, PAI-1 polymorphism, fibrinogen polymorphism, thrombomodulin polymorphisms, plasminogen deficiency). Hyperhomocysteinaemia is an example of inherited or acquired conditions that increase thrombotic risk due to direct toxic effects on endothelial cells as a result of increased tissue factor activity, increased platelet activation, suppression of thrombomodulin expression, and impaired fibrinolysis (58). Secondary hypercoagulable state Secondary hypercoagulable states generally occur as a result of a large number of transient or permanent acquired conditions that increase the tendency for formation of bloods because they convert constitutive non-thrombogenic surface of endothelial cells to the active proinflammatory thrombogenic phenotype. The conversion is a result of biochemical damage to endothelial cells mediated by several proinflammatory and prothrombotic mediators. This leads to increased expression of leukocyte adhesion molecules, increased expression of macrophages and tumour cell tissue factor, and inhibition of the protein C system. Endothelial conversion to the proinflammatory thrombogenic phenotype may be result of many transient and permanent disorders (57, 63, 64). Some of these such as the presence of antiphospholipd antibody, hyperhomocysteinaemia, polycythemia rubra vera, cancer and infection increase the risk of both arterial and venous thrombosis. Pregnancy, puerperium, the early postoperative state, sepsis, and oral contraceptives are examples of transient conditions where increased thrombotic risk is present as long as these transient disorders are present. Permanent disorders are the result of irreversible processes that produce alterations in basic homeostasis. Examples of these disorders are the antiphospholipid antibodies syndrome, carcinoma and other malignancies, nephrotic syndrome, previous thrombosis, myeloproliferative disorders and homocysteinaemia. The most common cause of an acquired hypercoagulable state is the antiphospholipid syndrome. Thrombotic events associated with the antiphospholipid syndrome (APS) more often affect the venous rather than arterial system. A heterogeneous family of antiphospholipid antibodies (anticardiolipin antibodies or lupus anticoagulans) are associated with many collagen vascular disorders, infections and some medications. The prevalence in the general population is 2% 5%, and it has also been described as an isolated abnormality. However, the incidence of APLA is in patients with SLE as high as 50% 60%. Although b 2 glycoprotein 1 is the most common target of APLAB, they have an affinity for the phospholipid surface of thrombin, high molecular weight kininogens, activated protein C and S and factor XII. Many

27 Alfirević and Alfirević: Conditions associated with hypercoagulable state S21 mechanisms have been proposed for the manner of how antiphospholid antibodies induce thrombosis. These range from endothelial defects to platelet and monocyte activation and alteration in plasma proteins. Many studies have confirmed the association between antiphospholipid antibodies and thrombosis (65 67). Cancer is the second most common acquired condition associated with the hypercoagulability state. It can lead to thrombus formation by local compression and invasion of the vessel wall, or indirectly with inflammatory and proangiogenic cytokines and tissue factor rich microparticles released from tumour cells. It is a well-recognized (AmE) risk factor for arterial and venous thrombosis (68). Use of some medications, such as aspariginase thalidomide, prothrombin complex/intermediate purity factor IX concentrate, oral contraceptives, and tamoxifen have shown a strong association with thrombosis (69). Oral contraceptives are an example of a well-defined condition associated with thrombosis (arterial and venous) as a result of endothelial dysfunction, changes in the concentrations of several coagulation factors and lipoprotein components. Oral contraceptives increase the risk of arterial thrombosis by approximately three-fold, and the risk for venous thrombosis by about four-fold (70). Many studies performed over the years have confirmed that acquired risk factors for venous thrombosis differ from those that cause arterial vascular disease. The latest study was a large LITE study in 2002 that included 19,293 people which corroborated the view that the aetiology of VTE differs from that of atherosclerotic cardiovascular disease (71). Hypertension, smoking, metabolic syndrome, and hyperlipidemia have been considered as risk factors for arterial thrombosis, whereas surgery, immobilisation, pregnancy and oestrogen use are the risk factors for venous thrombosis (71, 72). In the last few years, studies have shown that the pathophysiology of the hypercoagulable state has a complex mechanism and that risk factors for arterial and venous thrombosis that are traditionally divided, do overlap (73). In 2003, Prandoni was the first to report a higher prevalence of asymptomatic atherosclerosis lesions in patients with idiopathic deep vein thrombosis compared with patients with secondary DVT and controls (OR % CI ) (74). Additional comparative studies have shown that patients with arterial thrombosis (myocardial infarction and stroke) also had an increased relative risk of venous thrombosis in the first 3 months following the events mentioned above (75). In some patients, VTE might occur as the first symptomatic cardiovascular event (76, 77). Recently, a number of studies have found many common links between these two different clinical conditions that supported the finding that patients with venous thromboembolism and arterial thrombosis share common risk factors (78, 79). In addition to all these positive studies, it should also be mentioned that there are those who have not found any relationship (80, 81). The incidence of arterial and venous thrombosis increases exponentially with age due to increased activation of blood coagulation and fibrinolysis (82). Results of a meta-analysis by Ageno and colleagues from 2008 confirmed the hypothezis that cardiovascular risk factors for obesity (OR % CI ), hypertension (OR % CI ), diabetes mellitus (OR % CI ), smoking (OR % CI ) and hypercholesterolaemia (OR % CI ) are associated significantly with an increased risk of VTE (83). Patients with idiopathic VTE are at a higher-risk for complications of arterial thrombosis compared with those with VTE caused by well-known factors. Obesity is a well-known acquired risk for venous thrombosis due to immobility and venous stasis. In addition, abdominal obesity is a well-known component of the metabolic syndrome which is also an established risk factor for atheromatosis. In 2006, Ageno et al. first described the metabolic syndrome as an independent predictor of idiopathic DVT (OR % CI ) (84). In this study, patients with unexplained DVT had a significantly higher prevalence of the metabolic syndrome compared with controls. The patients with metabolic syndrome had increased platelet activity, increased plasma concentrations of PAI-1 and coagulation factors VIII, VII, XIII, and decreased production of nitric oxide and prostacyclin due to endothelial cell dysfunction. Procoagulant activity due to metabolic syndrome may act as a link between VTE and atherosclerosis. It is a result of the increased lipoprotein concentrations, increased amounts of circulating microparticles and increased proatherosclerotic mediators released from adipose cells (85). New results from the Copenhagen City Heart Study (2010) confirmed that obesity and smoking, two well-known cardiovascular risk factors, were both important risk factors for VTE body mass index whazard ratio (HR) for G35 vs. -20s2.10 (95% CI )x; smoking whr for G25 g tobacco per day vs. never smokers1.52 (95% CI )x; gender whr for men vs. womens1.24 (95% CI )x (86). This new perception suggested a closer link between the two different clinical conditions. However, the strength of this association is still controversial (87). A positive association between diabetes mellitus, dyslipidaemia with VTE and atherothrombosis can be explained by endothelial dysfunction, increased procoagulant concentrations and platelet activation and dysfibrinolysis. An improved understanding of inflammation, the role of lipoproteins, local hypercoagulability and endothelial injury integrates these processes, suggesting that venous and arterial thrombosis are two aspects of the same disease (73). The data indicate that inflammation of the vessel wall initiates thrombus formation, and inflammation and coagulation systems are coupled by a common activation pathway. According to current understanding, inflammation appears to be a common pathogenic determinant. Leukocytes are observed approximately 2 min after any type of endothelial cell injury and its proinflammatory mediators can further enhance thrombosis. It has been hypothesized that selections, expressed after a thrombogenic stimulus, facilitate these proposed interactions between leukocytes and endothelial cells, leukocytes and leukocytes, and leukocytes and platelets. Neutrophils and platelets then become activated, generating other mediators that amplify thrombosis (73, 88 90). Structural (increased col-

28 S22 Alfirević and Alfirević: Conditions associated with hypercoagulable state lagen and calcium content) and functional (degeneration of elastic fibrils) changes in the vessel wall during aging, or increased concentrations of plasma coagulation factors increase the prothrombotic risk (63). The formation of venous as well as arterial thrombosis is a result of complex gene-gene and gene-environment interactions involving many risk factors (72). Which clinical manifestations of thrombosis develop in genetically predisposed individuals is dependent on environmental and acquired risk factors. The most recent results from the Iowa Women s Health Study has confirmed this hypothesis. The risk for VTE was increased in elderly women who were smokers, physically inactive, overweight and diabetic, indicating that lifestyle contributes to risk of VTE (91). Epidemiology The prevalence of the hypercoagulable state depends on the ethnicity and clinical history of the population studied (63, 92). Inherited conditions are associated with venous thrombosis rather than arterial thrombosis (93). The incidence of venous thrombosis increases with age and number of inherited risk factors. In contrast, the influence of inherited risk factors in patients with arterial thrombosis declines with age. These individuals usually have more than one acquired factor rather than rare thrombophilic disorders (7, 63, 94). Testing for an inherited hypercoagulable state is successful in more than 60% of patients presenting with idiopathic venous thromboembolic disease, and informative in a very restricted population with arterial events (non-arteriosclerotic thromboembolic event in a young person or in population with premature occlusion after revascularization procedures) (92, 95). The association between inherited thrombophila and arterial thrombosis is weak (96). Mutations in factor II and FV Leiden are the two most common inherited causes of a hypercoagulable state in Caucasians. In Europe, the prevalence of FV Leiden and factor II varies across different geographical regions. The frequency of FV Leiden in Croatian healthy volunteers was 2.9%, in Greeks 7%, and Scandinavians 15% (59, 97 99). The prevalence is lowest in the general population, greater in individuals with a single thrombosis, and the highest in those with recurrent thrombosis or in family member who are thrombophilic (59). These mutations are rare in Asians and Africans and ethnic groups of Asian decent (Inuit Eskimos, Native Americans, Australian aborigines) (97, 98, 100). The prevalence of mutations in two natural anticoagulants, protein C and protein S, is higher in South Asians compared with Caucasians. These mutations were observed in up to 3% of Caucasians and in 6% of Southeast Asians with venous thromboembolism (100, 101). Asians also had the highest PAI-1 activity (102). Factor VIII is an important prothrombotic risk factor in the UK black population (levels )150 U/L; more than a 10-fold increased risk of VTE in blacks and is found in 35% of patients with VTE (102). The phenotypic frequency and genotypic frequency of PlA2 was estimated to be 26.5% and 15%, respectively, in individuals from Northern and central Europe (103). The antiphospholipid syndrome is the most common acquired condition associated with the hypercoagulable state. The risk for thrombosis increases with increased amounts antibody and lupus anticoagulants (104). Patients with more than one inherited thrombophilia defect are often at significantly greater risk of thrombosis than those Table 2 Prevalence of genetic disorders associated with the hypercoagulable state and relative risk for venous thrombosis. Hypercoagulable state Genetic disorders Frequency, % Relative risk Normals VTE patients Venous thrombosis Factor V Leiden (G1691A) qprothrombin G20210A 20 qoral contraceptives 30 Factor V Leiden (A1691A) Oral contraceptives 2 Prothrombin G20210A Protein C x Protein S x Antithrombin x Dysfibrinogenaemia Unknown Hyperhomocysteinaemia Factor VIII Factor IX Factor XI Lipoprotein (a) Thrombin-activatable fibrinolysis inhibitor (TAFI) Antiphospholipid antibody Lupus anticoagulant 11 Anticardiolipin antibody 3.1

29 Alfirević and Alfirević: Conditions associated with hypercoagulable state S23 with only a single genetic risk factor. The risk of thrombosis is usually expressed as relative risk and depends on the underlying cause of hypercoagulability. The prevalence of genetic disorders associated with the hypercoagulable state and relative risk for venous thrombosis is shown in Table 2 ( ). According to data from the literature, the relative risk of thrombosis is two times higher in patients with the prothrombin gene mutation, three times higher in patients with the mutation for FV Leiden, and six times higher in patients who are carriers of both mutations when compared to patients with the wild type genotype. The risk increases dramatically following exposure to some acquired conditions, such as oral contraceptives. FV Leiden and oral contraceptives raise the risk by 30-fold (108). Conclusions The hypercoagulable state is not a uniform disease, it is a complex condition that may or may not lead to thrombosis as a result of complex gene-gene and gene-environment interactions. Disorders of the hypercoagulable state due to thrombosis of veins and arteries are the most important cause of sickness and death in developed countries throughout the world today. It is well-known that tests for defects of inherited thrombophilia are not helpful in discovering the aetiology of arterial thrombosis, but are very useful in detecting venous thromboembolic disease. Inherited thrombophilia can be detected in more than 60% of patients with VTE. Screening tests for a primary hypercoagulable defect (inherited thrombophilia) are useful in individuals with thrombosis at an early age, a family history of thromboembolic disease, unusual sites of thrombosis, or recurrent thrombosis. There are no specific prophylactic therapies for any inherited conditions associated with the hypercoagulable state. Thrombosis associated with inherited thrombophilia is very often precipitated with some acquired conditions. Adequate identification of risk factors is essential in making decisions about the duration of anticoagulant therapy and prevention of recurrence (53, 108). Until recently risk factors for arterial and venous thrombosis have been considered to be strictly distinct. New epidemiological data and clinical trials have suggested that many acquired risk factors overlap and coexist for both disorders. Lifestyle changes, or timely recognition of acquired risk factors can reduce morbidity and mortality from these conditions. Conflict of interest statement Authors conflict of interest disclosure: The authors stated that there are no conflicts of interest regarding the publication of this article. Research funding: None declared. Employment or leadership: None declared. Honorarium: None declared. References 1. Schafer AI. The hypercoagulable states. Ann Intern Med 1985;102: Deitcher SR, Rodger GM. Thrombosis and antithrombotic therapy. In: Greer JP, John P, Foerster J, Rodgers G, Paraskevas M, Frixos G, Bertil A, Means DA, Robert T, editors. Wintrobe s clinical hematology, 12th ed. Philadelphia, PA: Lippincott Williams and Wilkins, 2009; Deitcher SR, Gomes MP. Hypercoagulable state testing and malignancy screening following venous thromboembolic events. Vasc Med 2003;8: Yusuf SH, Hawken S, Ounpuu S, Dent T, Avezum A, Lanas F, et al. On behalf of the INTERHEART Study Investigator. Effect of potentially modifiable risk factors associated with myocardial infarction in 52 countries (the INTERHEART study): case-control study. Lancet 2004;364: Badimon L, Badimon JJ, Galvez A, Chesebro JH, Fuster V. Influence of arterial wall damage and wall shear rate on platelet deposition: ex-vivo study in a swine model. Arteriosclerosis 1986;6: Jemal A, Siegel R, Ward E, Murray T, Xu J, Smigal C, et al. Cancer statistics, CA Cancer J Clin 2006;56: Hirsh H, Hoach J. Management of deep vein thrombosis and pulmonary embolism. A statement for health care professionals. Council on Thrombosis (in consultation with the Council on Cardiovascular Radiology), American Heart Association. Circulation 1996;93: Silverstein M, Heit J, Mohr D, Petterson T, O Fallon W, Melton L 3rd. Trends in the incidence of deep vein thrombosis and pulmonary embolism: a 25-year population-based, cohort study. Arch Intern Med 1998;158: Heit J, Petterson T, Farmer S, Bailey K, Melton L. Trends in incidence of deep vein thrombosis and pulmonary embolism: a 35-year population-based study. Blood 2006;108:430a. 10. Maekawa T. Hypercoagulable state and its clinical aspects. Nippon Ishikai Zasshi 1964;51: Egeberg O. Inherited antithrombin deficiency causing thrombophilia. Thromb Diath Haemorrh 1965;13: Deitcher SR, Caiola E, Jaffer A. Demystifying two common genetic predispositions to venous thrombosis. Cleve Clin J Med 2000;67: Furie B, Furie BC. Thrombus formation in vivo. J Clin Invest 2005;115: Kickler TS. Platelet biology an overview. Transfus Altern Transfus Med 2006;2: Monroe DM, Hoffman M. What does it take to make the perfect clot? Arterioscler Thromb Vasc Biol 2006;26: Levi M, ten Cate H, van der Poll T. Endothelium: interface between coagulation and inflammation. Crit Care Med 2002;30(Suppl 5):S Wright IS. The nomenclature of blood clotting factors. Can Med Assoc J 1962;86: Miletich JP, Jackson CM, Majerus PW. Properties of the factor Xa binding site on human platelets. J Biol Chem 1978;253: Prins MH, Hirsh J. A critical review of the evidence supporting a relationship between impaired fibrinolytic activity and venous thromboembolism. Arch Intern Med 1991;151: Ogawa S, Gerlach H, Esposito C, Pasagian-Macaulay A, Brett J, Stern D. Hypoxia modulates the barrier and coagulant function of cultured bovine endothelium. Increased monolayer per-

30 S24 Alfirević and Alfirević: Conditions associated with hypercoagulable state meability and induction of procoagulant properties. J Clin Invest 1990;85: Merli GJ. Pathophysiology of venous thrombosis and the diagnosis of deep vein thrombosis-pulmonary embolism in the elderly. Cardio Clin 2008;26: Gross PL, Aird WC. The endothelium and thrombosis. Tissue factor binds with factor VII and begins the coagulation cascade. Semin Thromb Hemost 2000;26: Coppens M, Kaandrop SP, Middeldrop S. Inherited thrombophilas. Obstet Gynecol Clin N Am 2006;33: Bell WR, Toby LS. Current status of pulmonary thromboembolic disease. Pathophysiology, diagnosis, prevention and treatment. Am Heart J 1982;103: Calvete JJ. On the structure and function of platelet integrin alpha IIb beta 3, the fibrinogen receptor. Proc Soc Exp Biol Med 1995;208: Rauch U, Osende JI, Fuster V, Badimon JJ, Fayad Z, Chesebro JH. Thrombus formation on atherosclerotic plaques: pathogenesis and clinical consequences. Ann Intern Med 2001; 34: Reiner AP, Siscovick DS, Roosendaal FR. Haemostatic pathogenesis of arterial thrombosis is result of multiple genetic (abnormalities of platelet function, fibrinogen level) and environmental factors (hypercholesterolemia, diabetes mellitus smoking) related to atherosclerosis and thrombosis. Risk factors and arterial thrombotic disease. Thromb Haemost 2001; 85: Wiliams MS, Bray PF. Genetics of arterial prothrombotic risk states. Exp Biol Med 2001;226:409 19T. 29. Bray PF, Howard TD, Vittinghoff E, Sane DC, Herrington DM. Effect of genetic variations in platelet glycoproteins Ibalpha and VI on the risk for coronary heart disease events in postmenopausal women taking hormone therapy. Blood 2007; 109: Weiss EJ, Bray PF, Tayback M, Schulman SP, Kickler TS, Becker LC, et al. A polymorphism of a platelet glycoprotein receptor as an inherited risk factor for coronary thrombosis. N Engl J Med 1996;334: Mammen EF. Ten years experience with the sticky platelet syndrome. Clin Appl Thromb Hemost 1995;1: Kannel WB, Wolf PA, Castelli WP, D Agostino RB. Fibrinogen and risk of cardiovascular disease. The Framingham Study. J Am Med Assoc 1987;258: Kanel WB. Hemostatic factors and the risk of myocardial infarction or sudden death in patients with angina pectoris. European Concerted Action on Thrombosis and Disabilities Angina Pectoris Study Group. N Engl J Med 1995;332: Meade TW. Fibrinogen in ischaemic heart disease. Eur Heart J 1995;16(Suppl A): Miller GJ, Bauer KA, Barzegar S, Cooper JA, Rosenberg RD. Increased activation of the haemostatic system in men at high risk of fatal coronary heart disease. Thromb Haemost 1996;75: Eriksson P, Kallin B, van t Hooft FM, Båvenholm P, Hamsten A. Allele-specific increase in basal transcription of the plasminogen-activator inhibitor 1 gene is associated with myocardial infarction. Proc Natl Acad Sci USA 1995;92: Thomas DP. Overview of venous thrombogenesis. Semin Thromb Hemost 1988;114: Becker BF, Heindl B, Kupatt C, Zahler S. Endothelial function and hemostasis. Z Kardiol 2000;89: Stewart GJ. Neutrophils and deep venous thrombosis. Haemostasis 1993;23: Prins MH, Hirsh JA. A critical review of the evidence supporting a relationship between impaired fibrinolytic activity and venous thromboembolism. Arch Intern Med 1991;15: Seligsohn U, Lubetsky A. Genetic susceptibility to venous thrombosis. N Engl J Med 2001;344: Rosendaal FR, Koster T, Vandenbroucke JP, Reitsma PH. High risk of thrombosis in patients homozygous for factor V Leiden (activated protein C resistance). Blood 1995;85: Koster T, Roosendaal FR, de Ronde H, Briet E, Vandenbroucke JP, Bertina RM. Venous thrombosis due to poor anticoagulant response to activated protein C: Leiden Thrombophilia Study. Lancet 1993;342: Poort SR, Roosendaal FR, Reitsma PH, Bertina RM. A common genetic variation in the 39-untranslated region of the prothrombin gene is associated with elevated plasma prothrombin levels and an increase in venous thrombosis. Blood 1996; 88: Margaglione M, Brancaccio V, Giuliani N, D Andrea G, Cappucci G, Lannaccone L, et al. Increased risk for venous thrombosis in carriers of the Prothrombin G A gene variant. Ann Intern Med 1998;129: Filip DJ, Eckstein JD, Veltkamp JJ. Hereditary antithrombin III deficiency and thromboembolic disease. Am J Hemat 1976;1: Khan S, Dickerman JD. Hereditary thrombophilia. Thromb J 2006;4: Rosenberg RD, Bauer KA. Thrombosis in inherited deficiencies of antithrombin, protein C and protein S. Hum Path 1987;18: Koster T, Blann AD, Briet E, Vandenbroucke JP, Rosendaal FR. Role of clotting factor VIII in effect of von Willebrand factor on occurrence of deep-vein thrombosis. Lancet 1995; 345: Meijers JC, Tekelenburg WL, Bouma BN, Bertina RM, Rosendaal FR. High levels of factor XI have been implicated in thrombosis. N Engl J Med 2000;342: Simioni P, Tormene D, Tognin G, Gavasso S, Bulato C, Iacobelli NP, et al. X-linked thrombophilia with a mutant factor IX (factor IX Padua). N Engl J Med 2009;36: Kirschsteln W, Simianer S, Dempfle CE. Impaired fibrinolytic capacity and tissue plasminogen activator release in patients with restenosis after PTCA. Thromb Haemost 1989;62: Rogers JS 2nd. Hypercoagulable states. W V Med J 1993; 89: Hassouna HI. Thrombophilia and hypercoagulability. Med Princ Pract 2009;18: Rao AK, Sheth S, Kaplan R. Inherited hypercoagulable states. Vasc Med 1997;2: Schafer AI, Levine MN, Konkle BA, Kearon C. Thrombotic disorders: diagnosis and treatment. Hematology Am Soc Hematol Educ Program 2003: Nachman RL, Silverstein R. Hypercoagulable states. Ann Intern Med 1993;119: Thomas RH. Hypercoagulability syndromes. Arch Intern Med 2001;161: Zöller B, García de Frutos P, Hillarp A, Dahlbäck B. Thrombophilia as a multigenic disease. Haematologica 1999;84: O Donnell J, Mumford AD, Manning RA, Laffan M. Elevation of FVIII: C in venous thromboembolism is persistent and independent of the acute phase response. Thromb Haemost 2000;83:10 3.

31 Alfirević and Alfirević: Conditions associated with hypercoagulable state S Patel RK, Ford E, Thumpston J, Arya R. Risk factors for venous thrombosis in the black population. Thromb Haemost 2003;90: Lowe GD. Common risk factors for both arterial and venous thrombosis. Br J Haemtol 2008;140: Chan MY, Andreoti F, Becker R. Hypercoagulable states in cardiovascular disease. Circulation 2008;118: Silvestrain RL. The vascular endothelium. In: Gallin JI, Snyderman R, editors. Inflammation: basic principles and clinical correlates. Philadelphia: Lippincott Williams and Wilkins, 1999: Love PE, Santoro SA. Antiphospholipid antibodies: anticardiolipin and the lupus anticoagulant in systemic lupus erythematosus (SLE) and in non-sle disorders. Prevalence and clinical significance. Ann Intern Med 1990;112: Shapiro SS. The lupus anticoagulant/antiphospholipid syndrome. Ann Rev Med 1996;47: Levine JS, Branch DW, Rauch J. The antiphospholipid syndrome. N Engl J Med 2002;346: Levine MN, Lee AY, Kakkar AK. From Trousseau to targeted therapy: new insights and innovations in thrombosis and cancer. J Thromb Haemost 2003;1: Cuzick J, Powles T, Veronesi U, Forbes J, Edwards R, Ashley S, et al. Overview of the main outcomes in breast-cancer prevention trials. Lancet 2003;25: Lowe GD. Update on the cardiovascular risks of hormone replacement therapy. Women s Health 2007;3: Tsai AW, Cushman M, Rosamond WD, Heckbert SR, Polak JF, Folsom AR. Cardiovascular risk factors and venous thromboembolism incidence: the longitudinal investigation of thromboembolism etiology. Arch Intern Med 2002;162: Voetsch B, Loscalzo J. Genetics of thrombophilia: impact on atherogenesis. Curr Opin Lipidol 2004;2: Martinelli I, Bucciarelli P, Mannucci PM. Thrombotic risk factors: basic pathophysiology. Crit Care Med 2010;38(Suppl 2):S Prandoni P, Bilora F, Marchiori A, Bernardi E, Petrobelli F, Lensing AW. An association between atherosclerosis and venous thrombosis. N Engl J Med 2003;348: Sørensen HT, Horvath-Puho E, Søgaard KK, Christensen S, Johnsen SP, Thomsen RW, et al. Arterial cardiovascular events, statins, low-dose aspirin and subsequent risk of venous thromboembolism: a population-based case-control study. J Thromb Haemost 2009;7: Becattini C, Agnelli G, Prandoni P, Silingardi M, Salvi R, Taliani MR, et al. A prospective study on cardiovascular events after acute pulmonary embolism. Eur Heart J 2005; 26: Prandoni P, Ghirarduzzi A, Prins MH, Pengo V, Davidson BL, Sørensen H, et al. Venous thromboembolism and the risk of subsequent symptomatic atherosclerosis. J Thromb Haemost 2006;4: Hong C, Zhu F, Du D, Pilgram TK, Sicard G, Bae K. Coronary artery calcification and risk factors for atherosclerosis in patients with VTE. Atherosclerosis 2005;183: Eliasson A, Bergqvist D, Björck M, Acosta S, Sternby NH, Ogren M. Incidence and risk of venous thromboembolism in patients with verified arterial thrombosis: a population study based on 23,796 consecutive autopsies. J Thromb Haemost 2006;4: van der Hagen PB, Folsom AR, Jenny NS, Heckbert SR, O Meara ES, Reich LM, et al. Subclinical atherosclerosis and the risk of future venous thrombosis in the Cardiovascular Health Study. J Thromb Haemost 2006;9: Reich LM, Folsom AR, Key NS, Boland LL, Heckbert SR, Rosamond WD, et al. Prospective study of subclinical atherosclerosis as a risk factor for venous thromboembolism. J Thromb Haemost 2006;9: Lowe GD, Rumley A, Woodward M, Morrison CE, Philippou H, Lane DA, et al. Epidemiology of coagulation factors, inhibitors and activation markers: the Third Glasgow MON- ICA Survey. I. Illustrative reference ranges by age, sex and hormone use. Br J Haematol 1997;97: Ageno W, Becattini C, Brighton T, Selby R, Kamphuisen W. Cardiovascular risk factors and venous thromboembolism: a meta-analysis. Circulation 2008;117: Ageno W, Prandoni P, Romuladi E, Ghirarduzzi A, Dentali F, Prsavento R, et al. The metabolic syndrome and the risk of venous thrombosis: a case-control study. J Thromb Haemost 2006;4: Piazza G, Goldhaber SZ. Venous thromboembolism and atheromathosis. Circulation 2010;121: Holst AG, Jensen G, Prescott E. Risk factors for venous thromboembolism: results from the Copenhagen City Heart Study. Circulation 2010;121: Prandoni P. Venous and arterial thrombosis. Two aspects of the same disease? Clin Epidemiol 2009;1: Stewart GJ. Neutrophils and deep venous thrombosis. Haemostasis 1993;23(Suppl 1): Falati S, Gross P, Merrill-Skoloff G, Furie BC, Furie B. Realtime in vivo imaging of platelets, tissue factor and fibrin during arterial thrombus formation in the mouse. Nat Med 2002;8: Schwertz H, Tolley ND, Foulks JM, Denis MM, Risenmay BW, Buerke M, et al. Signal-dependent splicing of tissue factor pre-mrna modulates the thrombogenicity of human platelets. J Exp Med 2006;203: Lutsey PL, Virnig BA, Durham SB, Steffen LM, Hirsch AT, Jacobs DR Jr, et al. Correlates and consequences of venous thromboembolism: The Iowa Women s Health Study. Am J Public Health 2010;8: Rosendaal FR. Venous thrombosis: a multicausal disease. Lancet 1999;353: Crowther A, Kelton JG. Congenital thrombophilic states associated with venous thrombosis: a qualitative overview and proposed classification system. Ann Intern Med 2003;138: Heit JA, Silverstein MD, Mohr DN, Tanya M, Petterson MS, O Fallon WM, et al. Risk factors for deep vein thrombosis and pulmonary embolism a population-based case-control study. Arch Intern Med 2000;160: de Moerloose P, Boehlen F. Inherited thrombophilia in arterial disease: a selective review. Semin Hematol 2007;44: Ye Z, Liu EH, Higgins JP, Keavney BD, Lowe GD, Collins R, et al. Seven hemostatic polymorphism and coronary disease: a meta analysis comprising 66,155 cases and 91,307 controls. Lancet 2006;367: Rees DC, Cox M, Clegg JB. World distribution of factor V Leiden. Lancet 1995;346: Roosendaal FR, Doggen C, Zivelin A, Arruda V, Aiach M. Geographic distribution of the 20210G to A prothrombin variant. Thromb Haemost 1998;79: Alfirevic Z, Simundic AM, Nikolac N, Sobocan N, Alfirevic I, Stefanovic M, et al. Frequency of factor II G20210A, factor V Leiden, MTHFR C677T and PAI-1 5G/4G polymorphism

32 S26 Alfirević and Alfirević: Conditions associated with hypercoagulable state in patients with venous thromboemblism: Croatian case-control study. Biochemia Medica 2010;20: Sakata T, Okamoto A, Mannami T, Matsuo H, Miyata T. Protein C and antithrombin deficiency are important risk factors for deep vein thrombosis in Japanese. J Thromb Haemost 2004;3: Franco RF, Reitsma PH. Genetic risk factors of venous thrombosis. Hum Genet 2001;109: Lutsey PL, Cushman M, Steffen LM, Green D, Barr RG, Herrington D, et al. Plasma hemostatic factors and endothelial markers in four racial/ethnic groups: the MESA study. J Thromb Haemost 2006;12: von dem Borne AE, Decary F. Nomenclature of platelet-specific antigens. Transfusion 1990;5: Bick RL, Baker WF. Antiphospholipid syndrome and thrombosis. Semin Thromb Hemost 1999;25: Whitlatch NL, Ortel TL. Thrombophilias: when should we test and how does it help? Sem Resp Crit Care M 2008;29: Lijfering WM, Veeger NJ, Brouwer JL, van der Meer J. The risk of venous and arterial thrombosis in hyperhomocysteinemic subjects may be a result of elevated factor VIII levels. Haematologica 2007;92: Martinelli I, Sacchi E, Landi G, Taioli E, Duca F, Mannucci PM. High risk of cerebral-vein thrombosis in carriers of a prothrombin-gene mutation and in users of oral contraceptives. N Engl J Med 1998;338: Rosendaal FR. Risk factors for venous thrombosis. Prevalence, risk and interaction. Semin Hematol 1997;34:

33 Clin Chem Lab Med 2010;48(Suppl 1):S27 S by Walter de Gruyter Berlin New York. DOI /CCLM Review Diagnostic algorithm for thrombophilia screening Sandra Margetic* Department of Laboratory Coagulation, University Department of Chemistry, Medical School University Hospital Sestre Milosrdnice, Zagreb, Croatia Abstract Thrombophilia screening is aimed at detecting the most frequent and well-defined causes of venous thrombosis, such as activated protein C resistance/factor V Leiden mutation, prothrombin G20210A gene mutation, deficiencies of natural anticoagulants, such as antithrombin, protein C and protein S, the presence of antiphospholipid antibodies, hyperhomocysteinemia and increased factor VIII activity. At this time, thrombophilia screening is not recommended for those possible congenital or acquired risk factors, whose association with increased risk of thrombosis has not been proven sufficiently. Laboratory investigations should include a stepwise approach to the diagnosis of thrombotic disorders with respect to the assays and methods of analysis that are used. The assays recommended for the first diagnostic step of screening should establish, whether the subject has one of the common causes of thrombophilia. If one or more abnormal results are obtained, the second diagnostic step includes the assays recommended for confirmation and/or characterization of the defect. When performing the investigation of thrombophilia, it is important to consider all pre-analytical and other variables that may affect the results of thrombophilia testing, including time of testing, age, gender, liver function, hormonal status, pregnancy or the acute phase response to inflammatory diseases. This is necessary, in order to avoid, any misinterpretation of the results. This review summarizes the current knowledge concerning thrombophilia investigations, with special focus on the diagnostic algorithm regarding patient selection, the assays and methods of analysis used and all the variables that should be considered when employing tests for the diagnosis of thrombophilia. Clin Chem Lab Med 2010;48:S Keywords: diagnostic algorithm; laboratory investigation; risk factors; thrombophilia; venous thromboembolism. *Corresponding author: Sandra Margetic, M.Sc., Department of Laboratory Coagulation, University Department of Chemistry, Medical School University Hospital Sestre Milosrdnice, Vinogradska 29, Zagreb, Croatia Phone: q , Fax: q , sandra.margetic1@zg.t-com.hr Received June 8, 2010; accepted September 5, 2010; previously published online November 6, 2010 Introduction The term thrombophilia is defined as the tendency to develop thrombosis due to pre-disposing factors that may be genetically determined, acquired, or both. Thrombosis may occur in both venous and arterial vessels, but thrombophilia is usually considered within the context of venous thromboembolism (VTE) (1 3). Although hereditary thrombophilia is associated primarily with VTE, it can also manifest with an arterial event specifically in the setting of a paradoxical embolism (4). Furthermore, some of the well-defined acquired risk factors for VTE, such as the presence of antiphospholipid antibodies (aplas) and hyperhomocysteinemia (HHC), can also present with arterial thrombosis. However, since at present there is no established causal relationship between hereditary thrombophilic risk factors and arterial thrombosis, testing for heritable thrombophilia is not indicated in patients with arterial thrombosis (5), but is directed primarily towards patients with VTE. Additionally, another indication for thrombophilia testing occurs in women with a history of adverse outcomes during pregnancy including severe pre-eclampsia, placental abruption, intrauterine growth restriction and unexplained consecutive first trimester abortions and second and third trimester unexplained fetal loss (6, 7). During the past few decades, the clinical importance of thromboembolic events has progressively increased. Today, VTE is a severe condition in all areas of medicine due to its endemic nature (8, 9). Each year, VTE affects approximately 1 2 individuals per 1000 in the general population of Western countries (10). Over the last two decades, knowledge on the etiology of VTE has increased considerably, and various hereditary and acquired risk factors have been discovered. As a consequence, the investigation of thrombophilia has contributed considerably to pressure on clinical laboratories, with enormously increased ordering of tests over the last 10 years. To date, well-defined risk factors for VTE include resistance to activated protein C (APCR)/factor V Leiden (FVL) mutation, prothrombin G20210A gene mutation (FIIG20210A), deficiencies in the natural anticoagulants antithrombin (AT), protein C (PC) and protein S (PS), the presence of aplas and HHC. In addition, since recent studies have clearly shown persistently increased concentrations of factor VIII (FVIII) to be an independent risk factor for VTE, it should be included in thrombophilia screening. In addition to these inherited and acquired risk factors for VTE, several environmental or transitional conditions, such as advancing age, malignant diseases, trauma, surgery, pregnancy and puerperium, immobilization and estrogen therapy, are asso- 2010/337

34 S28 Margetic: Thrombophilia diagnostic algorithm ciated with an increased risk for VTE (11 14). These conditions not only pre-dispose apparently healthy individuals to thrombosis but may also trigger thrombosis in persons with thrombophilic risk factors. Each individual risk factor constitutes an element of increased risk for VTE, which is compounded when several different conditions are present. Hence, individuals with multiple defects have considerably increased risk for VTE (15, 16). Furthermore, interactions between hereditary and transitional risk factors significantly increase the risk of VTE, suggesting the multifactorial nature of VTE (17). Thus, VTE often occurs in susceptible patients having one or more thrombophilic abnormalities when they are exposed to exogenous prothrombotic stimuli (18). However, single genetic or acquired risk factors do not necessarily lead to thrombosis without interaction with other genetic or environmental risk factors. When a thrombotic episode occurs, it is important to establish whether it represents an isolated event caused by a transitional condition, or whether the patient has an underlying genetic and/or acquired predisposition to life-long risk for thrombosis. At this time, the laboratory investigation of thrombophilia permits recognition of underlying inherited thrombotic risk factors in approximately 50% 60% of patients being evaluated (19). The aim of this review is to summarize the current knowledge concerning the laboratory diagnosis of thrombophilia. The review focuses on the most important aspects of laboratory investigation, with special interest on the diagnostic algorithm regarding careful patient selection, the assays and methods of analysis used, and all the pre-analytical variables and other conditions that should be considered when employing tests for the diagnosis of thrombophilia. General guidelines for laboratory investigation of thrombophilia Thrombophilia screening of an unselected population is not indicated. In general, the prevalence of any known risk factor for VTE is not sufficient to justify indiscriminate screening of the general population (20 23). Furthermore, the diagnostic investigation of thrombophilia is not indicated either in unselected patients presenting with a first episode of VTE, but is medically and economically justifiable in patients who have a history of unexplained thromboembolism. Therefore, all subjects with a confirmed episode of VTE should be considered for thrombophilia investigation if at least one of the following is present: thrombosis prior to the age of 50 years, recurrent spontaneous thrombosis, family history of VTE or thrombosis in unusual sites, such as portal, mesenteric, splenic, hepatic, renal or cerebral veins (24, 25), as shown in Figure 1. The initial investigation should begin with a complete personal and family medical history. Although a family history of thrombosis is a significant selection factor, it is important to note that a negative family history does not exclude a thrombophilic abnormality because the defects may have low penetrance and new mutations can occur (26). Because of the beneficial effect that thrombophilia screening may have on asymptomatic individuals, laboratory investigation is also justified for first degree family members of the symptomatic patient with previously identified inherited thrombophilic risk factors (27, 28). Thrombophilia screening in certain populations with a well-known increased risk for VTE, such as pregnant women and those on oral contraceptives is continually debated in the literature (29). Although indiscriminate testing of all pregnant women or those prior to prescription of oral contraceptives is not supported by the of majority investigations to date (5, 30), thrombophilia screening is indicated in selected patients with a previous history of VTE or who have a positive family history (5, 31). Furthermore, the results of some studies have shown that relatively weak inherited risk factors, such as heterozygosity for FVL and FIIG20210A mutations can be associated with spontaneous VTE in subjects )50 years of age (32, 33). This suggests that increased age acts as an additional risk factor (34). These results indicate that age at the time of VTE should not be taken as strict criteria to determine laboratory testing. The laboratory investigation of thrombophilia is aimed at detecting the common and well-established causes of thrombophilia in carefully selected patients with VTE. Thrombofilia testing should begin with global coagulation tests, such as the prothrombin time (PT), activated partial thromboplastin time (APTT), thrombin time (TT) and fibrinogen concentrations in order to rule out anticoagulant therapy and other coagulation disorders. The PT test indicates the effect of any anticoagulant treatment and functional status of the liver. The APTT and TT can reveal heparin therapy or contamination that interferes with some functional assays. Laboratory investigations should include a step-wise approach to the diagnosis of thrombotic disorders regarding the assays and methods of analysis used. As shown in Figure 1, the first diagnostic step should be restricted to the high priority tests for the most frequent and well-established causes of thrombophilia, including APCR/FVL, FIIG20210A mutation, deficiencies of natural anticoagulants AT, PC and PS, the presence of aplas, homocysteine (HC) and FVIII concentrations (35, 36). The general strategy of thrombophilia evaluation is to investigate individually each of these welldefined risk factors know to be associated with increased risk of thromboembolism. Since combined defects are common and their detection is clinically relevant because of the highly increased risk for thrombosis, a complete investigation should always be performed, even if one defect has already been identified. If none of the high priority test results are abnormal in patients with a family history of VTE or recurrent thrombosis, it is reasonable to perform low priority tests, such as the laboratory evaluation of dysfibrinogenemia (5, 37). Moreover, in the case of patients with unexplained intraabdominal thrombosis (portal, splenic, mesenteric or hepatic vein thrombosis), there is a need to consider molecular testing for Janus-activating kinase-2 (JAK-2) mutation to exclude an underlying myeloproliferative disorder and to evaluate CD55/CD59 expression by flow cytometry of peripheral blood to exclude paroxysmal nocturnal hemoglobinuria (PNH) (27, 38 40).

35 Margetic: Thrombophilia diagnostic algorithm S29 Figure 1 Diagnostic algorithm in the thrombophilia screening in patients presenting with VTE. At this time, the laboratory evaluation of thrombophilia is not recommended for other possible congenital or acquired risk factors, such as abnormalities of the fibrinolytic system (increased concentrations of plasminogen activator inhibitor-1, deficiencies of plasminogen or tissue plasminogen activator), deficiency of heparin cofactor II (HCII), increased concentrations of fibrinogen, FVII, FIX, FXI, or deficiency of FXII, whose association with an increased thrombotic risk has not been sufficiently proven at present (36, 41). The recommended assays in the first diagnostic step for screening should establish, whether the subject has one of the common causes of thrombophilia. If one or more abnormal results are obtained, a second diagnostic step should include recommended assays for the confirmation and/or characterization of the defect (Figure 1). The laboratory tests chosen in the investigation of thrombophilia should be specific and the results should be clinically relevant. Methodology issues in thrombophilia testing

36 S30 Margetic: Thrombophilia diagnostic algorithm are important because some methods for certain test are better than others. Hence, careful selection of specific recommended assays should be undertaken in order to ensure both test sensitivity and specificity (21, 42). Laboratory experts are in the best position to recognize the limitations of the tests they use, and they have an obligation to provide clinicians with both pre-test education and the limitations of the assays used. Laboratories should identify and report the most common pre-analytical causes of potential false-positive and false-negative results for all tests employed (2, 42). Also, quality assurance should be maintained through use of internal quality control as well as external quality assessment programmes in order to minimize errors (43 45). When performing the investigation of thrombophilia, it is particularly important to consider all of the pre-analytical and other variables that may affect the results of thrombophilia testing, including time of testing, age, gender, liver function, hormonal status, pregnancy or an acute phase response to inflammatory diseases in order to avoid misinterpretation of the results (21, 46). The appropriate time of testing is of critical importance because the acute phase of thrombosis and anticoagulant therapy considerably affect the results of many phenotypic assays, making interpretation of the results difficult (30, 46). Since the clinical management of an acute thrombotic event is not influenced by the immediate demonstration of a specific defect, the laboratory investigation should be delayed for 6 months after the acute phase of thrombosis, and at least 2 weeks after anticoagulant therapy has been discontinued. However, it is important to note that in some exceptional cases, immediate certain thrombophilic tests are indicated, such as testing for aplas in patients with catastrophic antiphospholipid syndrome (APLS) which may require more aggressive and/or prolonged therapy. Also, as discussed later in the section on natural anticoagulant deficiencies, in the case of patients with possible hereditary AT deficiency where the use of higher doses of heparin or AT concentrates may be life saving, determination of AT is indicated even in the acute phase of thrombosis and anticoagulant therapy. Furthermore, when evaluating pediatric patients for inherited thrombophilic risk factors, such as AT, PC and PS deficiencies, it is important to note that pediatric reference ranges are necessary because age-related changes occur in the concentrations of these proteins. It is recommended that appropriate age- and gender-dependent reference ranges for each parameter being evaluated be employed (2). Ideally, reference ranges should be determined for the method and test system being used locally, with the normal donors recruited from the local population. All pre-analytical and other variables that affect the results of thrombophilia testing will be discussed in the next section that describes the laboratory aspects in the evaluation of individual risk factors. In order to confirm any positive result in the investigation of thrombophilia that is obtained with non-genotyping tests, it is recommended that a second analysis be performed using a newly collected blood sample for any case with questionable or positive results (2, 21). Moreover, genetic defects can be definitely demonstrated through the investigation of first degree relatives. Laboratory aspects in the evaluation of individual risk factors in the thrombophilia screening Activated protein C resistance (APCR) and factor V Leiden (FVL) In 1993, Dahlback et al. first described a hereditary defect in the anticoagulant response to activated protein C (APC), known as APCR, that was associated with an increased risk for VTE (47). One year later, Bertina and colleagues at the University of Leiden identified the molecular basis of this defect (48). In over 90% of cases, inherited APCR is caused by a single point mutation in the factor V gene, known as FVL. The mutation results in inefficient inactivation of activated factor V (FVa) by APC, because the mutant FV is much more resistant to proteolytic degradation by APC. This mutation is the most common inherited risk factor for VTE in whites, with different prevalence in different populations and geographical regions. In Europe, the frequency of the FVL mutation is lowest in North Europe and highest in Southern Europe (49, 50). The FVL mutation is rarely found in African, Australian and South Asian populations (49). However, FVL accounts for most, but not all, cases of APCR. In the absence of FVL, an acquired APCR phenotype may be present during pregnancy, use of oral contraceptives, in the presence of lupus anticoagulants (LA) or with increased concentrations of FVIII, and in patients with myeloma who develop thrombosis (51 53). Recent studies have shown that acquired APCR is a risk factor for VTE, independent of FVL (51). The laboratory evaluation of APCR/FVL includes functional coagulation assays and genotyping for FVL by DNA analysis (54). Functional assays identify individuals who have resistance to APC, while DNA analysis identifies FVL mutation as the cause of the APCR. The most commonly used phenotypic screening test for APCR is based on prolongation of the APTT by the addition of APC, as originally described by Dahlback. Results are expressed as an APC ratio; the ratio between the APTT measured in the presence and then the absence of added APC. Although simple and inexpensive, this method is not sufficiently sensitive and specific for FVL. In the modified second generation of APTTbased methods for APCR, patient plasma is pre-diluted (1q4) with factor V-deficient plasma. This modification of the APCR assay is highly sensitive and specific for FVL mutation and can even be used in patients taking vitamin K antagonists (VKA) (27, 55). Standardization of the APTTbased APCR assay may be further improved by determination of a normalized ratio where the assay ratio is divided by the ratio of pooled normal plasma or standard human plasma analyzed in the same test run (42, 56). A general strategy to evaluate APCR/FVL is first-line screening by the APCR functional phenotypic test using the second generation APTT-based assay with FV-deficient plas-

37 Margetic: Thrombophilia diagnostic algorithm S31 ma. This test is automated, cost-effective and detects causes of APCR other than FVL. Borderline and positive results should undergo genotyping to confirm heterozygosity or homozygosity for FVL. A search for FVL by DNA analysis alone would not identify cases of APCR caused by acquired conditions. The variables that affect test results of phenotypic assays for APCR/FVL are shown in Table 1. Prothrombin G20210A mutation In 1996, Poort et al. described the mutation of the factor II (FII) gene which results in increased concentrations of functionally normal prothrombin in plasma due to increased synthesis (57). FIIG20210A mutation is the second most prevalent inherited risk factor for VTE, associated with a three-fold increase in risk. The mutation is present in approximately 2% of the general population and in 6% 7% of unselected patients with VTE (58). Although prothrombin activity is often raised mildly in carriers of the mutation, there is an overlap of values in subjects with and without mutation using the available functional test for FII activity. Therefore, the measurement of prothrombin activity in plasma is not an adequate test to screen thrombophilic patients for this mutation because it is unable to clearly distinguish carriers from non-carriers of the mutation (59). Thrombophilia investigations should include DNA analysis of the FIIG20210A mutation only (60), as shown in Figure 1. The acute phase of thrombosis and anticoagulant therapy do not affect DNA based assays. Deficiencies of natural anticoagulants Deficiencies of natural anticoagulants AT, PC and PS are less common genetic risk factors for thrombosis compared with APCR and FIIG20210A mutation, being present in -1% of the general population. Together they account for only 1% 5% of genetic defects found in patients with VTE (61, 62). AT deficiency was the first recognized inherited risk factor for VTE, described by Egeberg in 1965 (63). Later in the 1980s, deficiencies of PC and PS were identified as causes of inherited thrombophilia (64, 65). Since congenital deficiencies of natural anticoagulants are caused by a large number of different mutations, DNA testing is generally not available for thrombophilia investigations, outside of specialized research laboratories. Instead, the laboratory evaluation of natural anticoagulants includes functional and immunochemical assays. When considering a diagnosis of hereditary deficiencies of AT, PC and PS, it is of great importance to exclude the causes of acquired deficiencies since a large number of conditions are associated with a reduction in plasma concentrations of natural anticoagulants (Table 2). Also, there are numerous other variables that should be taken into consideration when employing thrombophilia assays for the natural anticoagulants, as shown in Table 2. However, in spite of a general recommendation to avoid thrombophilia testing for natural anticoagulants in patients on anticoagulant therapy and during the acute phase of thrombosis, in case of suspected hereditary AT deficiency, for example in patient with VTE from a family with known AT deficiency, measurement of AT is justified because it can modify therapy with higher doses of heparin, or even with AT concentrates. Further, in cases of suspected heparin resistance due to markedly decreased AT values, determination of AT is also justified because these patients may require very high doses of heparin in order to obtain an adequate therapeutic effect. However, in order to diagnose congenital AT deficiency, a decreased value of AT obtained during the acute thrombotic event or during heparin therapy must definitely be verified after the acute phase of thrombosis and anticoagulant therapy have been terminated. There are two major types of AT and PC deficiencies. Type I deficiency (quantitative) is caused by decreased synthesis of a biologically normal molecule, resulting in reductions in both the activity and concentration. Type II is a qualitative defect, resulting in decreased functional activity, but normal concentrations. In the first diagnostic step of laboratory investigation, the tests aimed at detecting AT and PC deficiencies should include functional assays capable of identifying both type I and type II abnormalities. Immunoassays are useful in combination with functional assays for the classification of the type of deficiency. If the result of functional assay is decreased, the immunochemical (antigenic) assay allows differentiation between type I and type II deficiencies (Figure 1). The immunochemical assay should not be performed without the functional assay because type II deficiencies in AT and PC will not be detected. Commercially available functional assays for AT are chromogenic methods in which AT activity can be measured according to its ability to inhibit thrombin or activated factor X (FXa). (66). Chromogenic methods with FXa are considered more appropriate compared with those based on thrombin as the target enzyme because they are not affected by the presence of other inhibitors of thrombin in plasma, i.e., HCII. On the basis of the nature of the functional defect, type II AT deficiency is subdivided into three subtypes depending Table 1 Variables that can affect the results of APCR/FVL phenotypic assays. Variable Impact on result Recommendation Contamination of plasma sample with platelets, False-positive result Double centrifugation of plasma sample is advised particularly in frozen and thawed samples if frozen samples are used Anticoagulant therapy with heparin, hirudin, False-positive result Do not perform testing in patients on therapy with argatroban these anticoagulants Heparin contamination False-positive result TT test can be used for excluding heparin contamination

38 S32 Margetic: Thrombophilia diagnostic algorithm Table 2 Variables that affect the test results for natural anticoagulants AT, PC and PS. Variables Impact on result Recommendation Anticoagulant therapy: AT activity decreases up to 30% during therapy with heparin Avoid testing in patients on anticoagulant therapy. heparin and vitamin K VKA cause notably reduced levels of PC and PS since both There are some exceptional cases in which AT measurement is antagonists (VKA) are vitamin K-dependent proteins needed in spite of acute phase of thrombosis or heparin therapy: patients with suspected congenital AT deficiency, as in case of patient with VTE from a family with known AT deficiency patients with suspected heparin resistance because of markedly reduced levels of AT In order to diagnose congenital AT deficiency in these cases, AT measurement must definitely be verified after the acute phase of thrombosis and anticoagulant therapy. Acute phase of thrombosis Decreased AT and PC because of increased consumption Avoid testing in the acute phase of thrombosis with the exception Lower concentrations of free PS because of increased of AT measurement in the cases listed above concerning to the C4B-BP as acute phase reactant anticoagulant therapy Age and gender At birth levels of AT, PC and PS are decreased until the age For pediatric patients, separated reference ranges should be of 6 months for AT and PS and until adolescence for PC constructed due to different levels of AT, PC and PS. PS in newborn is largely or entirely in the free form, because It is recommended to determine separate male and female reference C4B-BP is low or undetectable ranges of PS for adult population. Healthy women have slightly lower PS levels than men, and total PS concentrations increase with age in women Pregnancy and puerperium Decreased AT and PS Avoid testing: Estrogen therapy oral in pregnant women until 6 weeks postpartum contraceptives (OC) or in women taking oral contraceptives or HRT for three months hormone replacement with the exception of AT measurement in the cases listed above therapy (HRT) concerning to the anticoagulant therapy Acquired states of PC, PS AT: liver disease, sepsis, pre-eclampsia, nephrotic syndrome, pregnancy Acquired states of natural anticoagulants deficiencies should be and AT deficiencies and puerperium, surgery, post-traumatic state, therapy with L-asparaginase considered and excluded prior to thrombophilia investigation PC: DIC, acute thrombosis, sepsis, vitamin K deficiency, liver disease, postoperative state PS: DIC, acute thrombosis, vitamin K deficiency, liver disease,pregnancy and puerperium, nephrotic syndrome, inflammatory illness of any cause Clot-based functional Falsely low values of PC and PS may be obtained due to APCR/FVL Chromogenic functional assays are recommended as test of choice assays for PC and PS or FVIII)150%, and falsely increased values in the presence of LA or for PC in order to avoid interferences therapy with heparin or new anticoagulant drugs (hirudin, argatroban) Abnormal result of functional assay for PS should always be evaluated with an immunoassay for free PS Increased levels of C4B-BP Reduced level of free PS may be obtained because of In case of low PS activity it is advised to determine FVIII activity in inflammatory states increased binding to C4B-BP and a routine marker of acute phase (CRP, fibrinogen) Contamination of plasma Falsely low PS result Double centrifugation of plasma sample is advised if frozen sample with platelets, samples are used particularly in frozen and thawed samples VKA, vitamin K antagonists; DIC, disseminated intravascular coagulation; C4B-BP, C4B-binding protein; CRP, C-reactive protein.

39 Margetic: Thrombophilia diagnostic algorithm S33 Table 3 Classification of PS deficiencies on the basis of PS activity and free and total PS antigen concentrations. PS activity Free PS Total PS Type I Decreased Decreased Decreased Type II Decreased Normal Normal Type III Decreased Decreased Normal on whether the mutation affects the heparin binding site (HBS), reactive site (RS) or has pleiotropic (multiple) effects (PE) (56). Among these defects, HBS mutations are associated with a low-risk of thrombosis. This increases the clinical relevance of distinguishing the subtypes of AT deficiencies (67). Crossed immunoelectrophoresis, with and without heparin, or a variant of functional chromogenic assay that measures progressive inhibitory activity may be used to identify subtypes of type II AT deficiencies (56, 67). Commercially available functional assays for PC are either clot-based or chromogenic methods (42, 56). Chromogenic methods are generally recommended because clot-based assays are affected by several interferences, such as the presence of LA, increased FVIII concentrations ()150%) and APCR (Table 2). Although at present, there is no evidence that distinguishing between type I and type II PC deficiency is clinical relevant, immunochemical PC assays are often routinely employed in cases of reduced functional activity, since it can serve as a useful additional quality control step by comparison with functional PC levels (42). Immunochemical assays for AT and PC measurements include enzyme-linked immunosorbent assays (ELISAs) and newer automated immunoturbidimetric assays. The laboratory investigation of PS deficiency is more complex than that for AT and PC because of the lack of well-standardized functional assays. There are three types of congenital PS deficiencies on the basis of PS activity and free and total PS antigen concentrations (Table 3). Type I and type III PS deficiency are quantitative defects with both low free PS antigen concentrations and low PS activity, and account for 95% of cases (68). However, type I deficiency is associated with low total PS antigen, whereas type III deficiency is associated with normal PS antigen. PS type II deficiency accounts for approximately 5% of cases, representing a qualitative defect where concentrations of both total and free PS antigen are normal, but PS activity is diminished (68). The laboratory evaluation of PS deficiency should include a functional assay and immunological assay for PS antigen (69). A functional assay can be used as the initial screening assay because it detects all three types of deficiency. If only antigenic assays are performed in the first diagnostic step, type II deficiencies will not be detected. Functional assays for PS are clot-based methods that are widely used for initial screening of thrombophilia. These methods measure the ability of PS to serve as a cofactor to APC, enhancing its degradation of activated FV and FVIII, thereby prolonging clotting time. However, the limitations of these methods regarding interferences of LA, increased FVIII concentrations and APCR should be taken into consideration (42, 70), as shown in Table 2. Therefore, abnormal (low) result for PS obtained using functional assay should be further investigated using immunochemical assays (5, 71). Immunochemical methods, including ELISA and newer automated immunoturbidimetric assays, can measure total and the free PS concentrations, depending on the assay design. Free PS antigen assays measure only unbound or free PS, while total antigen assays measure bound (inactive) and unbound (free and active) PS. Namely, in plasma PS is present in two forms: 60% of entire PS is in a complex with the C4B-binding protein (C4B-BP), while 40% is present in an free form, which is the active component of PS. In order to determine the free antigen, in older methods, bound PS is precipitated using polyethylene glycol (PEG), followed by measurement of the remaining PS portion in the supernatant. Newer methods use specific monoclonal antibodies for free PS, which makes the precipitation step unnecessary. It is generally considered that free PS antigen, being closely related to the functional form of PS, shows better discrimination between subjects with and without PS deficiency compared with total PS concentrations (69, 72). If free PS antigen is decreased only, a total antigen assay can be performed to further determine the deficiency subtype (type I and type III), although it is not necessary to routinely measure total PS antigen (72). In addition to the functional and immunochemical assays for PC and PS natural anticoagulants described above, a commercial global screening test for the PC pathway was developed as an attempt for the global testing of the entire potential of the PC anticoagulant pathway (35, 73). Defects that affect the PC pathway are PC or PS deficiencies and APCR. However, these global tests for PC anticoagulant system have shown acceptable diagnostic efficacy only for APCR and PC deficiency, but not for PS deficiency where variable proportions of patients with this deficiency were not identified using the global test (74, 75). Therefore, at this time, the available PC pathway functional assays cannot be proposed as a global test for the PC anticoagulant system in thrombophilia screening because of insufficient sensitivity (76). Antiphospholipid antibodies (aplas) APLS, characterized by the presence of circulating aplas, represents an important cause of acquired thrombophilia associated with both VTE and arterial thromboembolism, recurrent fetal loss and thrombocytopenia (77). Diagnosis of APLS requires demonstration of persistently increased concentrations of aplas in the setting of thrombosis or certain complications of pregnancy (78). The aplas are a heterogenous group of autoantibodies directed against the phospholipid-protein complexes. Currently, no single test is sufficient for the diagnosis of aplas because of the lack of specific tests, and the heterogeneity of the antibodies (79, 80). Based primarily on the method of detection, there are three major subgroups of aplas: LA, anticardiolipin anti-

40 S34 Margetic: Thrombophilia diagnostic algorithm bodies (ACL) and anti-b 2 -glycoprotein-1 (anti-b 2 GP1) antibodies. LA antibodies are identified by functional coagulation assays, where they prolong clotting times. In contrast, ACL and anti-b 2 GP1 antibodies are detected by immunoassays that measure the immunological reactivity to a phospholipid or a phospholipid-binding protein. In accordance with recommended guidelines, the laboratory investigation of aplas should always include determination of all three groups of antibodies (78, 81). Lupus anticoagulants In vitro, LA prolong various clotting times because of binding to phospholipids where they interfere with the ability of phospholipids to serve its essential cofactor function in the coagulation cascade. According to the International Society of Thrombosis and Haemostasis guidelines, laboratory detection of LA consists of a panel of coagulation-based assays that include 1. prolongation of at least one phospholipid-dependent coagulation test, 2. evidence of inhibitor activity in the test plasma determined by mixing tests with pooled normal plasma, and 3. confirmation that the inhibitory effect is due to blocking of phospholipiddependent coagulation by the addition of excess phospholipids (82). Therefore, patients may be considered to be LA positive if one or more phospholipid-dependent tests are prolonged, if the prolonged test is not corrected when plasma from the patient and a healthy control are mixed and if prolongation of the test is corrected by increasing the concentration of phospholipids in the test system. There is no definite recommendation on the assays of choice for LA testing. Because no one single screening test is 100% sensitive for LA, it is advised that two or more screening tests with different assay principles be used (83, 84). The most common used combination of LA screening tests includes the APTT, the mixing test and dilute Russell s viper venom time (drvvt) test; the later of which is considered to be the most sensitive. The test is performed with Russell viper venom that directly activates coagulation factor X leading to the formation of a fibrin clot. LA prolongs the drvvt clotting time by interfering with phospholipids. The most common used confirmatory LA test is drvvt which contains an excess of phospholipids that neutralize the effect of LA resulting in correction of the drvvt screening test (85). Variables that affect the results of coagulation assays for LA are presented in Table 4. Anticardiolipin antibodies and anti-b 2 -glycoprotein-1 antibodies The laboratory evaluation of ACL and antib 2 GP1 antibodies is performed using ELISA methods (86). Both isotypes IgG and IgM of ACL and anti-b 2 GP1 antibodies should be measured. Generally, ACL antibodies are sensitive, but not particularly specific, while anti-b 2 GP1 antibodies have greater specificity for APLS (87 89). Testing for the presence of anti-b 2 GP1 has been shown to be very useful in patients that have persistently low titers of aplas. Patients who are anti-b 2 GP1 positive should be considered as having aplas. The laboratory criteria for the diagnosis of aplas should be positive on two or more separate occasions performed at least 12 weeks apart. This is of particular importance because the transient occurrence of aplas, which is sometimes observed in conjunction with microbial infection or drugs, is not associated with an increased risk of thrombosis. Among the variables affecting the results of ACL, it is important to note that false-positive IgM ACL result can be associated with rheumatoid factor and cryoglobulins (90). Hyperhomocysteinemia (HHC) HHC, or increased plasma HC concentrations have been shown to be associated with an increased risk for VTE, as well as for arterial thrombosis (91 94). HHC may be caused by a congenital deficiency of enzymes, such as cystathionine-b-synthetase (CBS) and methylenetetrahydrofolate reductase (MTHFR) involved in its metabolism, or may be attributable to poor dietary intake of vitamins B6, B12 or folate that act as cofactors in HC metabolism (36). Homozygous deficiency of CBS, a rare hereditary defect with a prevalence in the general population only of 0.3%, is characterized by increased plasma concentrations of HC, the presence of HC in the urine, atherosclerosis, and arterial and venous thrombosis occurring at a young age. In contrast, mutation of the MTHFR gene appears in homozygous form in 10% 13%, and in heterozygous form in 30% 40%, of the general population (95). A common variant in the gene for MTHFR that results in a thermolabile variant of the enzyme has been shown to be associated with mildly increased HC concentrations. However, recent studies have indicated that MTHFR polymorphism does not predispose to HHC when Table 4 Variables that affect the results of coagulation assays for the detection of lupus anticoagulants. Variables Impact on result Recommendation Plasma sample with platelet count False-negative LA result because of Double centrifugation of plasma )10=10 9 /L neutralization of LA sample is advised in order to avoid the interference of platelets Acute phase of thrombosis False-negative LA result because of Do not perform testing in the acute possible consumption phase of thrombosis Anticoagulant therapy: heparin and False-positive result for LA screening Do not perform testing in patients new anticoagulant drugs (hirudin, tests (APTT, mixing tests) on any anticoagulant therapy argatroban, danaparoid, bivalirudin) Vitamin K antagonists (VKAs) False-positive LA result FVIII activity)150% False-negative results for LA FVIII activity may be measured in screening tests (APTT, mixing tests) order to exclude acute phase reaction

41 Margetic: Thrombophilia diagnostic algorithm S35 folate status is adequate. Only when combined with vitamin deficiencies, heterozygous subjects often have mildly increased HC concentrations. Most heterozygotes do not experience HHC or increased risk for thrombosis, unless they have other thrombotic risk factors. The recent MEGA Study, a large population-based case-control study by Bezemer and collaborators, has clearly shown that there is no association between the MTHFR 677CT polymorphism and risk for VTE (96). Therefore, genetic testing of the MTHFR polymorphism is not indicated in thrombophilia screening, as it has not been shown to be risk factor for thrombosis if HC concentrations are normal (97, 98). The laboratory diagnosis of HHC as a part of the investigation of thrombophilia should be performed exclusively by determining a fasting HC concentration in plasma, without genotyping for genetic defects. Until recently, HC concentrations have been measured primarily using high-pressure liquid chromatography (HPLC). Now, newer and simpler immunoassays are commercially available. These methods are more suited for use in the general clinical laboratory because they require simpler instrumentation and less technical expertise compared with HPLC (99). The acute phase of thrombosis and anticoagulant therapy do not affect the results of HHC investigations. EDTA or citrated plasma may be used, but plasma should be separated from cells within an hour of collection to avoid contamination with HC from red cells. Controlled blood collection and proper sample handling for HC measurement is essential to obtain valid results. Testing for HC concentrations should be done following an overnight fast of 10 h before collection of blood. The diet should be normal and not supplemented with vitamins in the few weeks preceding testing. Immediately after collection, the blood sample should be placed on ice or refrigerated at 2 88C until analysis. Elevated factor VIII Increased FVIII concentrations were first associated with increased thrombotic risk by Koster et al. in 1995, as part of the Leiden Thrombophilia Study (100). This study has demonstrated that FVIII values )150% are an independent risk factor for VTE, with a relative risk of three-fold and a highrisk of recurrence. These findings were confirmed by O Donnell et al. who found increased FVIII to be the common abnormality in patients referred for thrombophilia screening because of unexplained VTE (101). To date, no genetic variation in the FVIII gene has been identified that might account for this variation in phenotype. However, in 1998, Kamphuisen observed a positive correlation of FVIII concentrations within families (102, 103). Although the molecular basis of an increased FVIII is not presently known, the high reported prevalence of increased FVIII concentrations suggest that a genetic component may be at least partly responsible for factor increases. Among patients with VTE, the prevalence of increased plasma FVIII concentrations is approximately 20% 25%, and high FVIII persists over time and appears to be independent of the acute phase response (104). The significant number of patients with isolated persistent increased FVIII concentrations as their only risk factor, as well as the increased risk of recurrent VTE in these patients suggests that routine screening for thrombophilia should also include measurement of this coagulation factor (105, 106). For thrombophilia screening purposes, assays that measure both the activity and antigen of FVIII are suitable. The activity of FVIII can be measured using APTT-based methods with FVIII-deficient plasma or by chromogenic methods. The antigen can be measured using ELISA-based methods. The most widely used methods are clot-based methods. Measurements of FVIII concentrations should be delayed for at least 6 months after an acute thrombotic event, since FVIII is an acute phase reactant. Furthermore, since FVIII concentrations are increased physiologically during pregnancy and the puerperium period, thrombophilia testing should be delayed for at least 6 weeks postpartum. In cases with increased FVIII concentrations as a result of thrombophilia screening, it is advised to repeat the testing 3 6 months after initial testing in order to confirm FVIII increases. Dysfibrinogenemia Dysfibrinogenemias are a very rare, but heterogenous group of congenital disorders resulting in a structurally altered fibrinogen molecule that may cause altered fibrinogen function (107). These disorders may be clinically asymptomatic or may cause bleeding and venous or arterial thrombosis (108). The prevalence of the disorder in patients with VTE is 0.8% (108). Laboratory investigation of dysfibrinogenemias should include simple screening assays, such as TT and reptilase time (RT). These are typically prolonged in dysfibrinogenemias because of a dysfunctional fibrinogen molecule. Positive cases identified using TT and RT assays should be evaluated by parallel analysis of functional and immunoreactive fibrinogen. Functional fibrinogen assays show considerably lower concentrations compared with immunological assays that measure fibrinogen quantity. This is because fibrinogen function is impaired, but the quantity of fibrinogen is not. A decreased fibrinogen activity/antigen ratio is considered the confirmatory test for the diagnosis of dysfibrinogenemia. The most commonly used functional activity assay for fibrinogen is the Clauss method, while immunoreactive fibrinogen may be determined by ELISA methods or by newer automated immunoturbidimetric methods. As dysfibrinogenemia is very uncommon risk factor for VTE, laboratory evaluation of this disorder is not performed as a part of routine thrombophilia screening, but only if an initial panel of high priority tests excludes more common causes of VTE (5, 109), as shown in Figure 1. Since fibrinogen is an acute phase reactant, laboratory evaluation of dysfibrinogenemia should be delayed 6 months after acute thrombotic event.

42 S36 Margetic: Thrombophilia diagnostic algorithm Conclusions and perspectives The laboratory investigation of thrombophilia should be performed in accordance with the recommended evidence-based guidelines on testing, regarding who and when to evaluate and what tests and test methods to use. Following these guidelines is necessary in order to ensure that an accurate diagnosis is made. Excessive or inappropriate thrombophilia testing outside the recommended guidelines is likely to be more harmful than beneficial for the patient due to the possibility of misinterpretation of the test results. In addition, because of the lack of global tests for screening thrombophilic patients, the laboratory investigation of thrombophilia requires an expensive approach by performing a panel of different assays for each patient being evaluated. Future efforts should be aimed at developing a diagnostic strategy that makes testing more cost-effective by developing new screening tests able to assess globally the increased tendency for VTE. Since a substantial proportion of patients have no identifiable cause of thrombosis, future efforts should focus attention on the identification of additional risk factors for VTE. Also, studies in progress should investigate the relevance of interactions between different genetic risk factors, or interactions between genetic and acquired risk factors in the development of VTE. More knowledge of risk factors and particularly of their inter-relationships would help to improve our understanding of the mechanisms of thrombus formation in a variety of conditions, as well as enable us to individualize risk and guidelines for thrombosis prevention. Conflict of interest statement Authors conflict of interest disclosure: The authors stated that there are no conflicts of interest regarding the publication of this article. Research funding: None declared. Employment or leadership: None declared. Honorarium: None declared. References 1. Tripodi A. Laboratory diagnosis of thrombophilic states: where do we stand? Pathophysiol Haemost Thromb 2002;32: Favaloro EJ, McDonald D, Lippi G. Laboratory investigation of thrombophilia: the good, the bad and the ugly. Semin Thromb Hemost 2009;35: Bauer KA, Rosendaal FR, Heit JA. Hypercoagulability: too many tests, to much conflicting data. Hematol 2002;1: Carano N, Agnetti A, Hagler DJ, Tchana B, Squarcia U, Bernasconi S. Acute myocardial infarction in a child: possible pathogenic role of patent foramen ovale associated with heritable thrombophilia. Pediatrics 2004;114: Baglin T, Gray E, Greaves M, Hunt BJ, Keeling D, Machin S, et al. Clinical guidelines for testing for heritable thrombophilia. Br J Haematol 2010;149: Robertson R, Wu O, Greer I. Thrombophilia and adverse pregnancy outcome. Curr Opin Obstet Gynecol 2004;16: Hossain N, Shamsi T, Soomro N. Frequency of thrombophilia in patients with adverse pregnancy outcome. J Pak Med Assoc 2005;55: Cohen AT, Agnelli G, Anderson F, Arcelus JI, Bergqvist D, Brecht JG, et al. Venous thromboembolism (VTE) in Europe. The number of VTE events and associated morbidity and mortality. Thromb Haemost 2007;98: Bronić A. Thromboembolic diseases as biological and clinical syndrome Role of the Mediterranean League against Thromboembolic Diseases. Biochem Med 2010;20: Spek CA, Reitsma PH. Genetic risk factors for venous thrombosis. Mol Genet Metab 2000;71: Falanga A. Thrombophilia in cancer. Semin Thromb Haemost 2005;31: Allman-Farinelli MA, Dawson B. Diet and aging: bearing on thrombosis and hemostasis. Semin Thromb Hemost 2005;31: James AH, Jamison MG, Brancazio LR, Myers ER. Venous thromboembolism during pregnancy and the postpartum period: incidence, risk factors and mortality. Am J Obstet Gynecol 2006;194: Hoffman R, Brenner B. Thrombophilia related issues in women and children. Semin Thromb Hemost 2005;31: Lippi G, Franchini M. Pathogenesis of venous thromboembolism: when the cup runneth over. Semin Thromb Hemost 2008;34: Brouwer JL, Veeger NJ, Kluin-Nelemans HC, van der Meer J. The pathogenesis of venous thromboembolism: evidence of multiple interrelated causes. Ann Intern Med 2006;145: Rosendaal FR. Venous thrombosis: the role of genes, environment, and behaviour. Hematology 2005;1: Wu O, Robertson L, Twaddle S, Lowe GD, Clark P, Greaves M, et al. Screening for thrombophilia in high-risk situations: systematic review and cost-effectiveness analysis. The Thrombosis: Risk and Economic Assessment of Thrombophilia Screening (TREATS) study. Health Technol Assess 2006;10: Cohn DM, Roshani S, Middeldorp S. Thrombophilia and venous thromboembolism: implications of testing. Semin Thromb Hemost 2007;33: Cumming AM, Shiach CR. The investigation and management of inherited thrombophilia. Clin Lab Haem 1999;21: Carraro P. European Communities Confederation of Clinical Chemistry and Laboratory Medicine (EC4) Working Group on Guidelines for Investigation of Disease. Guidelines for the laboratory investigation of inherited thrombophilias. Recommendations for the first level clinical laboratories. Clin Chem Lab Med 2003;41: Mannucci PM. Laboratory detection of inherited thrombophilia: a historical perspective. Semin Thromb Hemost 2005;31: Dalen JE. Should patients with venous thromboembolism be screened for thrombophilia? Am J Med 2008;121: College of American Pathologists Consensus Conference XXXVI: Diagnostic issues in thrombophilia. Arch Pathol Lab Med 2002;126: Nicolaides AN, Breddin HK, Carpenter P, Coccheri S, Conard J, De Stefano V, et al. European Genetics Foundation Cardiovascular Disease Educational and Research Trust International Union of Angiology Mediterranean League on Thromboembolism. Thrombophilia and venous thromboembolism. International consensus statement. Guidelines according to scientific evidence. Int Angiol 2005;24:1 26.

43 Margetic: Thrombophilia diagnostic algorithm S De Stefano V, Finazzi G, Mannucci PM. Inherited thrombophilia: pathogenesis, clinical syndromes and management. Blood 1996;87: Moll S. Testing for thrombophilias. Clin Adv Hematol Oncol 2008;6: Tripodi A. Issues concerning the laboratory investigation of inherited thrombophilia. Mol Diagn 2005;9: Scifres CM, Macones GA. The utility of thrombophilia testing in pregnant women with thrombosis: fact or fiction? Am J Obstet Gynecol 2008;199: Middeldorp SS, van Hylckama Vileg A. Does thrombophilia testing help in the clinical management of patients? Br J Haematol 2008;143: Lindhoff-Last E, Luxembourg B. Evidence based indications for thrombophilia screening. Vasa 2008;37: Ridker PM, Glynn RJ, Miletich JP, Goldhaber SZ, Stampfer MJ, Hennekens CH. Age-specific incidence rates of venous thromboembolism among heterozygous carriers of factor V Leiden mutation. Ann Intern Med 1997;126: De Stefano V, Rossi E, Paciaroni K, D Orazio A, Cina G, Marchitelli E, et al. Different circumstances of the first venous thromboembolism among younger or older heterozygous carriers of the G20210A polymorphism in the prothrombin gene. Haematologica 2003;88: Rosendaal FR, van Hylckama VA, Doggen CJ. Venous thrombosis in the elderly. J Thromb Haemost 2007;5(Suppl 1): Tripodi A, Mannucci MP. Laboratory investigation of thrombophilia. Clin Chem 2001;47: Robetorye RS, Rodgers GM. Update on selected inherited venous thrombotic disorders. Am J Hematol 2001;68: Seligsohn U, Lubetsky A. Genetic susceptibility to venous thrombosis. N Engl J Med 2001;344: McMahon C, Abu-Elmagd K, Bontempo FA, Kant JA, Swerdlow SH. JAK2 V617F mutation in patients with catastrophic intra-abdominal thromboses. Am J Clin Pathol 2007;127: Wautrecht JC. Venous thromboembolic disease: which coagulation screening, for whom, when? Rev Med Brux 2005;26: Parker C, Omine M, Richards S, Nishimura J, Bessler M, Ware R, et al. Diagnosis and management of paroxysmal nocturnal hemoglobinuria. Blood 2005;106: Crowther MA, Roberts J, Roberts R, Johnston M, Stevens P, Skingley P, et al. Fibrinolytic variables in patients with recurrent venous thrombosis: a prospective cohort study. Thromb Haemost 2001;85: Jennings I, Cooper P. Screening for thrombophilia: a laboratory perspective. Br J Biomed Sci 2003;60: Favaloro EJ. Learning from peer assessment: the role of the external quality assurance multilaboratory thrombophilia test process. Semin Thromb Hemost 2005;31: Plebani M, Sanzari MC, Zardo L. Quality control in coagulation testing. Semin Thromb Hemost 2008;34: Bonar R, Favaloro EJ, Adcock DM. Quality in coagulation and haemostasis testing. Biochem Med 2010;20: Favaloro EJ, Lippi G, Adcock DM. Preanalytical and postanalytical variables: the leading causes of diagnostic error in hemostasis? Semin Thromb Hemost 2008;34: Dahlback B, Carlsson M, Svensson PJ. Familial thrombophilia due to a previously unrecognized mechanism characterized by poor anticoagulant response to activated protein C. Proc Natl Acad Sci USA 1993;90: Bertina RM, Koeleman BP, Koster T, Rosendaal FR, Dirven RJ, de Ronde H, et al. Mutation in blood coagulation factor V associated with resistance to activated protein C. Nature 1994;369: Rees DC, Cox M, Clegg JB. World distribution of factor V Leiden. Lancet 1995;346: Lucotte G, Mercier G. Population genetics of factor V Leiden in Europe. Blood Cells Mol Dis 2001;27: Graf LL, Welsh CH, Qamar Z, Marlar RA. Activated protein C resistance assay detects thrombotic risk factors other than factor V Leiden. Am J Clin Pathol 2003;119: Remkova A. Diagnostic approach to hypercoagulable states. Bratisl Lek Listy 2006;107: Zangari M, Saghafifar F, Anaissie E, Badros A, Desikan R, Fassas A, et al. Activated protein C resistance in the absence of factor V Leiden mutation is a common finding in multiple myeloma and is associated with an increased risk of thrombotic complications. Blood Coag Fibrinolysis 2002;13: Press RD, Bauer KA, Kujovich JL, Heit JA. Clinical utility of factor V Leiden (R506Q) testing for the diagnosis and management of thromboembolic disorders. Arch Pathol Lab Med 2002;126: Svensson PJ, Zoller B, Dahlback B. Evaluation of original and modified APC-resistance tests in unselected outpatients with clinically suspected thrombosis and in healthy controls. Thromb Haemost 1997;77: Khor B, van Cott EM. Laboratory evaluation of hypercoagulability. Clin Lab Med 2009;29: Poort SR, Rosendaal FR, Reitsma PH, Bertina RM. A common genetic variation in the 39-untranslated region of the prothrombin gene is associated with elevated plasma prothrombin levels and an increase in venous thrombosis. Blood 1996;88: Zoller B, Garcia de Frutos P, Hillarp A, Dahlback B. Thrombophilia as a multigenic disease. Haematologica 1999;84: Grunewald M, Germowitz A, Beneke H, Guethner C, Griesshammer M. Coagulation II activity determination is not useful as a screening tool for the G20210A prothrombin gene allele (letter). Thromb Haemost 2000;84: McGlennen RC, Key NS. Clinical and laboratory management of the prothrombin G20210A mutation. Arch Pathol Lab Med 2002;126: Coppola A, Tufano A, Cerbone AM, Di Minno G. Inherited thrombophilia: implications for prevention and treatment of venous thromboembolism. Semin Thromb Hemost 2009;35: Rosendaal FR, Reitsma PH. Genetics of venous thrombosis. J Thromb Haemost 2009;7(Suppl 1): Egeberg O. Inherited antithrombin deficiency causing thrombophilia. Thromb Diath Haemorrh 1965;13: Griffin JH, Evatt B, Zimmerman TS, Kleiss AJ, Wideman C. Deficiency of protein C in congenital thrombotic disease. J Clin Invest 1981;68: Comp PC, Esmon CT. Recurrent venous thromboembolism in a patients with a partial deficiency of protein S. N Engl J Med 1984;311: Kottke-Marchant K, Duncan A. Antithrombin deficiency: issues in laboratory diagnosis. Arch Pathol Lab Med 2002;126: Rossi E, Chiusolo P, Za T, Marietti S, Ciminello A, Leone G, et al. Report of a novel kindred with antithrombin heparin-

44 S38 Margetic: Thrombophilia diagnostic algorithm binding site variant (47 Arg to His): demand for an automated progressive antithrombin assay to detect molecular variants with low thrombotic risk. Thromb Haemost 2007;98: Castoldi E, Hackeng TM. Regulation of coagulation by protein S. Curr Opin Hematol 2008;15: Van Cott EM, Ledford-Kraemer M, Meijer P, Nichols WL, Johnson SM, Perrschke EI. Protein S assays. Am J Clin Pathol 2005;123: Walker ID, Greaves M, Preston FE. Investigation and management of heritable thrombophilia. Br J Haematol 2001;114: Haemostasis and Thrombosis Task Force, British Committee for Standards in Haematology. Investigation and management of heritable thrombophilia. Br J Haematol 2001;114: Goodwin AJ, Rosendaal FR, Kottke-Marchant K, Bovill EG. A review of the technical, diagnostic, and epidemiologic considerations for protein S assays. Arch Pathol Lab Med 2002; 126: Robert A, Eschwege V, Hameg H, Drouet L, Aillaud MF. Anticoagulant response to Agkistrodon contortrix venom (AVC test): a new global test to screen for defects in the anticoagulant protein C pathway. Thromb Haemost 1996;75: Rimmer JE, Cooper PC, Brookfield CJ, Preston FE, Makris M. Evaluation of a global screening assay for the investigation of the protein C anticoagulant pathway. Clin Lab Haematol 2000;22: Gemmati D, Serino ML, Tognazzo S, Ongaro A, Moratelli S, Gilli G, et al. The reduced sensitivity of the ProC Global test in protein S deficient subjects reflects a reduction in the associated thrombotic risk. Blood Coagul Fibrin 2001;12: Gardiner C, Cooper PC, Makris M, Mackie IJ, Malia RG, Machin SJ. An evaluation of screening tests for defects in the protein C pathway: commercial kits lack sensitivity and specificity. Blood Coagul Fibrin 2002;13: Gezer S. Antiphospholipid syndrome. Dis Mon 2003;49: Miyakis S, Lockshin MD, Atsumi T, Branch DW, Brey RL, Cervera R, et al. International consensus statement on an update of the classification criteria for definite antiphospholipid syndrome (APS). J Thromb Haemost 2006;4: Wisloff F, Jacobsen EM, Liestol S. Laboratory diagnosis of the antiphospholipid syndrome. Thromb Res 2002;108: Favaloro EJ, Wong RC. Laboratory testing and identification of antiphospholipid antibodies and the antiphospholipid syndrome: a potpourri of problems, a compilation of possible solutions. Semin Thromb Hemost 2008;34: Galli M. Clinical utility of laboratory tests used to identify antiphospholipid antibodies and to diagnose the antiphospholipid syndrome. Semin Thromb Hemost 2008;34: Brandt JT, Triplett DA, Alving B, Scharrer IM. Criteria for the diagnosis of lupus anticoagulants: an update. Thromb Haemost 1995;74: Levine JS, Branch WD, Rauch J. The antiphospholipid syndrome. N Engl J Med 2002;346: Pengo V, Tripodi A, Reber G, Rand JG, Ortel TL, Galli M, et al. Update of the guidelines for lupus anticoagulant detection. J Thromb Haemost 2009;7: Triplett DA. Use of the dilute Russell viper venom time (d- RVVT): its importance and pitfalls. J Autoimm 2000;15: Triplett DA. Antiphospholipid antibodies. Arch Pathol Lab Med 2002;126: Galli M, Luciani D, Bertolini G, Barbui T. Anti-beta-2-glycoprotein I, antiprothrombin antibodies, and the risk of thrombosis in the antiphospholipid syndrome. Blood 2003;102: Reber G, de Moerloose P. Anti-beta-2-glycoprotein I antibodies when and how should they be measured? Thromb Res 2004;114: Marai I, Gillburd B, Blank M, Shoenfeld Y. Anti-cardiolipin and anti-beta2-glycoprotein I (beta2gp-i) antibody assays as screening for anti-phospholipid syndrome. Hum Antibodies 2003;12: Dodig S. Interferences in quantitative immunochemical methods. Biochem Med 2009;19: Edirisinghe SP. Homocysteine-induced thrombosis. Br J Biomed Sci 2004;61: Falcon CR, Cattaneo M, Penzeri D, Martinelli I, Mannucci PM. High prevalence of hyperhomocysteinemia in patients with juvenile venous thrombosis. Arterioscler Thromb 2000; 14: Makris M. Hyperhomocysteinemia and thrombosis. Clin Lab Haem 2000;22: Cattaneo M. Hyperhomocysteinemia and venous thromboembolism. Semin Thromb Hemost 2006;32: Guba S, Fonseca V, Link L. Hyperhomocysteinemia and thrombosis. Semin Thromb Hemost 1999;253: Bezemer ID, Doggen CJ, Vos HL, Rosendaal FR. No association between the common MTHFR 677C-)T polymorphism and venous thrombosis: results from the MEGA study. Arch Intern Med 2007;167: Key NS, McGlennen RC. Hyperhomocyst(e)inemia and thrombophilia. Arch Pathol Lab Med 2002;126: Alfirevic Z, Simundic AM, Nikolac N, Sobocan N, Alfirevic I, Stefanovic M, et al. Frequency of factor II G20210A, factor V Leiden, MTHFR C677T and PAI-1 5G/4G polymorphism in patients with venous thromboembolism: croatian case-control study. Biochem Med 2010;20: Tripodi A, Chantarangkul V, Lombardi R, Lecchi A, Mannucci PM, Cattaneo M. Multicenter study of homocysteine measurement. Performance characteristics of different methods, influence of standards on interlaboratory agreement of results. Thromb Haemost 2001;85: Koster T, Blann AD, Briet E, Vanderbroucke JP, Rosendaal FR. Role of clotting factor VIII in effect of von Willebrand factor on occurrence of deep-vein thrombosis. Lancet 1995; 345: O Donnell J, Mumford AD, Manning RA, Laffan M. Elevation of FVIII: C in venous thromboembolism is persistent and independent of the acute phase response. Thromb Haemost 2000;83: Kamphuisen PW, Houwing-Duistermaat JJ, van Houwelingen HC, Eikenboom JC, Bertina RM, Rosendaal FR. Familial clustering of factor VIII and von Willebrand factor levels. Thromb Haemost 1998;79: Bank I, Libourel EJ, Middeldorp S, Hamulyak K, van Pampus EC, Koopman MM, et al. Elevated levels of FVIII:C within families are associated with an increased risk for venous and arterial thrombosis. J Thromb Haemost 2005;3: Kraaijenhagen RA, in t Anker PS, Koopman MM, Reitsma PH, Prins MH, van den Ende A, et al. High plasma concentration of factor VIIIc is a major risk factor for venous thromboembolism. Thromb Haemost 2000;83: Kyrle PA, Minar E, Hirschl M, Bialonczyk C, Stain M, Schneider B, et al. High plasma levels of factor VIII and the risk of recurrent venous thromboembolism. N Engl J Med 2000;343: Benjaber K, Constans J, Cougnard A, Salmi LR. High levels

45 Margetic: Thrombophilia diagnostic algorithm S39 of factor VIIIc and risk of venous thrombosis: critical analysis of case control studies. Rev Med Intern 2003;24: Cunningham MT, Brandt JT, Laposata M, Olson JD. Laboratory diagnosis of dysfibrinogenemia. Arch Pathol Lab Med 2002;126: Haverkate F, Samama M. Familial dysfibrinogenemia and thrombophilia. Report on a study of the SSC subcommittee on fibrinogen. Thromb Haemost 1995;73: Hayes T. Dysfibrinogenemia and thrombosis. Arch Pathol Lab Med 2002;126:

46

47 Clin Chem Lab Med 2010;48(Suppl 1):S41 S by Walter de Gruyter Berlin New York. DOI /CCLM Review Genetic basis of thrombosis Valeria Bafunno 1 and Maurizio Margaglione 1,2, * 1 Genetica Medica, Dipartimento di Scienze Biomediche, Università degli Studi di Foggia, Foggia, Italy 2 Unità di Emostasi e Trombosi, I.R.C.C.S. Casa Sollievo della Sofferenza, S. Giovanni Rotondo, Italy Abstract Venous thrombosis (VT) represents a common and serious disorder that occurs as the result of clotting of the blood in the venous system and venous obstruction. Environmental risk factors and genetic predisposition play an important role in the development of thrombosis. It is therefore seen as a classic example of a complex common disease. We have focused on the role of genetic risk factors, primarily related to the hemostatic system, in triggering thrombotic events. Since the identification of antithrombin deficiency in 1965, major efforts have been made during the past 15 years to identify other genetic entities that lead to increased thrombotic risk. Results of early genetic studies demonstrated that two types of genetic defects cause VT: loss of function mutations in the natural anticoagulants antithrombin, protein C and protein S and gain of function mutations in procoagulant factors V (FV Leiden) and II (prothrombin G20210A). The high incidence of these mutations in Caucasians induced a shift from family studies to case-control association studies. Several investigations have been performed on the role of other candidate genetic risk factors predisposing to VT, including such variants in FXIII, FIX and fibrinogen genes. Moreover, the contribution of genetic variation in genes encoding less-well studied proteins that are part of the anticoagulant pathways has been evaluated. Recently, different genome-wide association studies have been performed in which several single nucleotide polymorphisms were investigated and related to the risk of VT. However, further studies are needed to identify additional genetic causes of thrombosis and to assess functional molecular mechanisms. Clin Chem Lab Med 2010;48:S *Corresponding author: Maurizio Margaglione, MD, Istituto di Genetica Medica, Dipartimento di Scienze Biomediche, Università degli Studi di Foggia, Viale Pinto, Foggia 71100, Italy Phone: q , Fax: q , m.margaglione@unifg.it Received July 6, 2010; accepted September 5, 2010; previously published online October 30, 2010 Keywords: case-control studies; family studies; genetic risks factors; venous thrombosis. Introduction Thrombosis has been described as hemostasis in the wrong place (1). Genetic studies in thrombosis started with coining of the term thrombophilia by Jordan and Nandorff in Since then, researchers have attempted to elucidate the role of genetic factors in venous thrombosis (VT) (2). VT, whose main clinical presentations include deep vein thrombosis (DVT), usually in the leg, and pulmonary embolism (PE), represents a common and serious disorder with an age dependent incidence of 1 3 per 1000 individuals per year (3). It is still the major cause of morbidity and mortality during pregnancy and childbirth in developed countries, and is a frequent cause of disease in young women using oral contraceptives. Both sexes are equally afflicted by a firsttime VT, but the risk of recurrent thrombosis is higher in men than in women (3, 4). One of the first organizations founded to bring together basic scientists and specialists, clinicians interested in different field related to thrombosis was the Mediterranean League against Thromboembolic Diseases. The goal of this organization was promoting research into the pathogenesis of thromboembolic diseases, prevention, diagnosis and treatment, and thus of providing overall better care of patients (5). There is no doubt that the concept of common complex disease is pertinent in this clinical condition: VT may be seen as a consequence of a heterogeneous collection of genetic and environmental factors. Although the role of environmental factors (such as trauma and immobilization, surgery, pregnancy, use of oral contraceptives) in triggering thrombotic events have captured the public s interest, major advances have been made over the past 15 years in understanding the genetics of VT. This review focuses on the role of genetic risk factors, mainly related to the haemostatic system, in triggering thrombotic events. Data on each of the well-established genetic conditions predisposing to VT are summarized, and the contribution of other candidate genetic risk factors is discussed. Data from family based studies of the genetics of thrombosis in Spanish individuals suggest that more than 60% of the variation in susceptibility to common thrombosis is attributable to genetic components (6). Reports describing families with a marked predisposition to VT were published as early as in 1905 (7). Subsequently, hundreds of mutations 2010/398

48 S42 Bafunno and Margaglione: The role of genetic factors in venous thrombosis in different genes have been identified, enabling an understanding of the etiology and genetic complexity of VT. Genetic risks factors for VT may be classified as strong, moderate and weak. Deficiencies in the natural anticoagulants, antithrombin (AT), protein C (PC) and protein S (PS) are strong risk factors (risks 5 10-fold increased). Moderate risk factors are Factor V Leiden and prothrombin 20210A (risk increase 2 5-fold). Weak genetic risks factors include variants that increase risk no more than 1.5-fold (8). A large population-based case-control study showed that a positive family history increased the risk of VT more than 2-fold wor, 2.2 (95% CI, )x and up to four-fold w3.9 (95% CI, )x when more than one relative was affected. Therefore, a positive family history increases susceptibility to VT. For those with a genetic and environmental risk factor, and a positive family history, the risk was about 64- fold higher than for those with no known risk factors and a negative family history (9). Deficiencies of natural anticoagulation inhibitors Deficiencies in the group of proteins usually referred to as natural coagulation inhibitors, i.e., AT, PC and PS are considered rare disorders and are found in -1% of the general population; prevalence rates for AT deficiency of 1 in 500 to 1 in 5000 in the overall population have been reported (10). AT is a serine protease inhibitor that physiologically inactivates thrombin and factor Xa and, to a lesser extent, factors IXa, XIa, XIIa. AT deficiency was first reported in 1965 by Egeberg, who described a Scandinavian family in which several subjects with decreased plasma AT concentrations presented with thrombotic events. Since then, numerous studies have described similar clinical conditions in additional families, establishing the concept of AT deficiency as a risk factor for thrombosis (11). AT deficiency is inherited as an autosomal dominant trait. Most cases are heterozygous; homozygosity for AT deficiency is rare and is almost always fatal in utero. Heterozygosity for AT deficiency can be found in 4% of families with inherited thrombophilia, in 1% of consecutive patients with a first DVT, and in 0.02% of healthy individuals (12). AT deficiency is divided into type I (low plasma concentrations of AT) and type II (dysfunctional protein in plasma). Type I AT deficiency is much more prevalent, often representing up to 80% of total cases. Type II is further subdivided into three types depending on the location of the mutation: IIa (defective reactive site), IIb (defective heparin-binding site) and IIc (pleiotropic group of mutations) (13). The gene AT3 coding for AT is located on chromosome 1q23 25, spanning 13.4 kb with 7 exons that encode for a mature secreted peptide of 432 amino acids (14). The molecular basis of AT deficiency is highly heterogeneous. Following the identification of the first mutation in 1983, the analysis of the AT3 gene has revealed a myriad of mutations (11, 15). The most frequent genetic defects are missense mutations, but other types of gene lesions, such as nonsense mutations, splice site mutations, deletions and insertions, have also been reported (13). The current online AT3 gene database contains 256 entries (13) and a total of 127 different mutations have been reported. Type I AT deficiencies are most commonly caused by short deletions and insertions, and less commonly by single base pair substitutions, especially nonsense mutations. Three preferred regions for these deletions have been identified, codons 244/245, codon 81 and codons 106/107. The type II AT deficiencies are commonly secondary to single base pair substitutions. Most mutations involve the reactive domain occurring at residues Ala382 and Ala384 in the hinge region of AT, and near the reactive domain at residues 392, 393 and 394. Most mutations altering the heparin binding capacity of AT are missense mutations, often involving residues 41, 47, 99 and 129. The type IIc deficiencies include a host of different mutations, especially involving residues 402, and 429. These mutations are located in strand Ic of AT, impairing the function of the reactive domain and also giving rise to a reduction in AT concentrations. This decrease is because strand Ic is necessary for the structural and functional integrity of AT (16). The estimates of thrombotic risk linked to AT deficiency and its prevalence in thrombotic patients also vary between different investigations, probably reflecting differences in study design and patient selection. In general, data obtained from family studies indicate increased risks (17, 18). PC and PS deficiencies result in defects in the activated PC anticoagulant system. PC, a vitamin K-dependent plasma glycoprotein, is cleaved to its activated form, activated protein C (APC), and then acts as a serine protease to inactivate factors Va and VIIIa. In the 1980s, genetic defects leading to PC and PS deficiencies were for the first time reported as causes of inherited VT (19, 20). PC deficiency is inherited as an autosomal dominant disorder, and heterozygosity for PC deficiency is a significant risk factor for VT. However, autosomal recessive inheritance has been observed in families with severe thrombosis resulting from homozygous or compound heterozygous PC deficiency (21). Heterozygosity for PC deficiency is found in 6% of families with inherited thrombophilia, in 3% of patients with a first DVT, and in 0.3% of healthy individuals (12). The PROC gene encoding for PC has been localized to chromosome 2q13 14 where it spans approximately 10 kb and contains 9 exons. Similar to AT, PC deficiency is classified into type I (low plasma concentration of both functional and immunological PC) and type II (low plasma concentrations of functional protein with normal antigen concentrations). Mutations in the PC gene resulting in deficiency of PC are highly heterogeneous (22, 23). The mutation database for PC deficiency in its 1995 edition contained 331 entries (160 different) for 315 apparently unrelated probands from 16 different European and American countries (24). The majority of these entries are missense, nonsense and splicing mutations. PS is a potent anticoagulant protein that down regulates thrombin formation by stimulating the proteolytic inactiva-

49 Bafunno and Margaglione: The role of genetic factors in venous thrombosis S43 tion of factors Va and FVIIIa by APC. The PROS1 gene encoding for PS has been mapped to 3p11.1, spanning 80 kb and containing 15 exons. The inheritance pattern of familial PS deficiency is usually autosomal dominant. PS deficiency is classified as type I (quantitative deficiency of both total and free PS), type II (qualitative deficiency characterized by decreased activity and normal total and free PS antigen concentrations) and type III (normal concentrations of total PS and low concentrations of free PS). At present, more than 200 mutations have been described in the PROS1 gene, and large deletions/duplications have been identified as being relatively common causes of PS deficiency (25 27). Because AT, PC and PS deficiencies are so rare, most reports on the prevalence and thrombotic risk of heterozygotes come from family studies (17, 21, 28, 29). Heterozygous carriers of AT, PC and PS deficiencies show a highly penetrant phenotype, and similar thrombotic risks 10-fold higher than seen in non-carriers. However, the risks conferred by these deficiencies appear lower in unselected consecutive patients with thrombosis than in the families with penetrant thrombophilia (28, 30). This is probably because these thrombophilic families present with additional risks for thrombosis, such as resistance to APC or other unknown genetic defect (31, 32). Bucciarelli et al. performed a retrospective cohort family study to assess the risk for venous thromboembolism (VTE) in individuals with AT, PC, or PS deficiency, or activated PC resistance (APCR). The lifetime risk for VTE was 4.4 for AT vs. APCR, 2.6 for AT vs. PS, 2.2 for AT vs. PC, 1.9 for PC vs. APCR, and 1.6 for PS vs. APCR. AT deficiency seems to have a higher risk for VTE compared with the other genetic defects (18). In 2008, another retrospective family cohort study was performed in order to assess whether hereditary PS, PC, or AT deficiency was associated with arterial thromboembolism (ATE). The authors found that deficient subjects had a 4.7- fold (95% CI, ; ps0.007) higher risk for ATE before the age of 55 years vs. 1.1 (95% CI, ) thereafter, compared with non-deficient family members. For separate deficiencies, the risks were 4.6- (95% CI, ), 6.9- (95% CI, ), and 1.1- (95% CI, ) fold higher in PS-, PC-, and AT-deficient subjects, respectively, before 55 years of age (33). Factor V Leiden In 1993, Dahlbäck and colleagues described an inherited abnormality that was highly prevalent in patients with VT. This condition was referred to as APCR defined as a poor anticoagulant response of plasma to the addition of APC (34). The phenotype of APC resistance is, in most cases, the result of a gain-of-function point mutation in the coagulation factor V gene: a G A transition at nucleotide position 1691 leading to substitution of Arg 506 by Gln in the FV molecule (FV Leiden), one of the three cleavage sites for APC. FV Leiden is the most common genetic defect involved in the etiology of VT. In Europeans, this mutation is found in up to 20% of familial thromboses. The transmission is autosomal dominant (1q21) (35). The extent of the increased thrombotic risk linked to FV Leiden is probably less than that associated with AT, PC and PS deficiencies. Martinelli et al. made a direct comparison of the thrombotic risk associated with deficiency of anticoagulant proteins and FV Leiden in 150 Italian families with different thrombophilic defects. They found higher risks for thrombosis in subjects with AT wrisk ratio 8.1 (95% CI, )x, PC w7.3 (95% CI, )x, PS deficiency w8.5 (95% CI, )x, and FV Leiden w2.2 (95% CI, )x, compared with individuals with normal coagulation. The risk of thrombosis for subjects with FV Leiden was lower than that for those with all three other coagulation defects, even when the analysis was restricted to DVT w0.3 (95% CI, )x (36). However, data obtained in case-control and cohort studies show that the relative risk for VT is about 3 10 for heterozygotes and for homozygotes (35, 37). The prevalence of FV Leiden carriers in the healthy population varies between different geographical regions and is clearly population dependent. The FV:Q506 allele has a high frequency in Caucasians (the prevalence of heterozygotes varies from 2% to 13%), and is extremely rare in indigenous populations of Asia, Africa, America and Australia. This population-based distribution may explain the higher incidence of thromboembolism in Caucasians compared with other groups (38). Haplotype analysis of individuals homozygous for the FV:Q506 allele shows strong evidence for a founder effect, with the single mutational event occurring 21,000 34,000 years ago (39, 40). Lindqvist et al. demonstrated that women carrying the FV:Q506 allele have a significantly lower risk of intrapartum bleeding complications compared with non-carriers, providing evidence that the mutation has conferred selective advantages in Caucasian populations (41). Alfirevic et al. performed a prospective study to assess the prevalence of FV Leiden, FII, MTHFR and plasminogen activator inhibitor (PAI)-1 polymorphisms in Croatian patients with thromboembolic disease. Of the polymorphisms tested, only the FV Leiden polymorphism showed to be significantly associated with VTE, with an odds ratio OR (95% CI)s6.41 ( ); ps0.004 (42). Whether FV Leiden is also associated with an increased risk of recurrent VTE remains controversial (43 45). In 1997, Bernardi et al. identified a FV haplotype, denoted HR2, as a group of six polymorphisms in the FV gene associated with a decreased APC response (46). This was probably related either to the amino acid substitutions predicted by the HR2 haplotype, or to linkage disequilibrium with another mutation in the FV gene. However, conflicting results about the relationships between the HR2 haplotype and the risk of thrombosis have been obtained in different studies. Other mutations affecting other APC cleavage sites in FVa have been described including FV Cambridge, FV:R306T and FV Hong Kong, FV:R306G that could potentially give rise to APC resistance, hypercoagulability and an increased risk of VT (47, 48). These gene variations are rare in the general population, and it remains to be demonstrated

50 S44 Bafunno and Margaglione: The role of genetic factors in venous thrombosis whether FV Hong Kong and FV Cambridge are a risk factor for thrombosis. Prothrombin 20210A A G-to-A transition at nucleotide of the prothrombin gene has been identified as the second most common independent risk factor for VT (49). The mutation is located in the 39UTR of the gene and is associated with increased plasma concentrations of prothrombin (approx. 30% in heterozygotes and 70% in homozygotes) which may be the molecular mechanism behind the increased risk of thrombosis. The transmission is also autosomal dominant (11p11). Data obtained from the Leiden Thrombophilia Study (LETS) have established that the prothrombin A allele was associated with an increased risk of a first DVT wor 2.8, (95% CI, )x. Its distribution is elective in Caucasians, with a North to South gradient in Europe (prevalence of and 0.03, respectively) (50). This gain of function mutation was found in 1% 3% of subjects in the general population, approximately in 6% of unselected patients with VT, and in 10% of probands of thrombophilic families. As previously reported, it remains uncertain if individuals homozygous or double heterozygous for FV Leiden and/or prothrombin G20210A have an increased risk of recurrent VT. A recent case-control study calculated the risk of recurrent VT in individuals with homozygosity or double heterozygosity of FV Leiden and/or prothrombin G20210A. The OR for recurrence was 1.2 (95% CI, ) for heterozygous carriers of FV Leiden, 0.7 (95% CI, ) for prothrombin G20210A, 1.2 (95% CI, ) for homozygous carriers of FV Leiden and/or prothrombin G20210A, and 1.0 (95% CI, ) for double heterozygotes of both mutations. Adjustments for age, gender, family status, first event type, and concomitance of natural anticoagulant deficiencies did not alter the risk estimates. Therefore, individuals homozygous or doubly heterozygous for FV Leiden did not have a high-risk of recurrent VT (51). Hyperhomocysteinemia Hyperhomocysteinemia is an abnormal increase in plasma concentrations of homocysteine, and is associated with a mild (approx. 2- to 4-fold) increase in thrombotic risk (52). Genetic and acquired factors interact to determine plasma homocysteine concentrations. Genetic causes include gene defects in methylenetetrahydrofolate reductase (MTHFR) and cystathionine b-synthase (CBS); two enzymes involved in homocysteine intracellular metabolism that may result in enzyme deficiency and hyperhomocysteinemia. Most of the mutations in MTHFR and CBS genes are rare and only have clinical consequence in homozygotes or compound heterozygotes. This condition, known as homocystinuria, is an autosomal recessive disorder characterized by multiple neurological deficits, psychomotor retardation, seizures, skeletal abnormalities, lens dislocation, premature arterial disease and VTE (53). Several mutations in MTHFR and CBS have been identified in patients with homocystinuria (54 56). However, this review we will focus on two mutations in MTHFR (C677T and A1298C) and one mutation in CBS (844ins68). Individuals with compound heterozygous or homozygous MTHFR polymorphisms may have slight increases in homocysteine concentrations, especially when folate deficiency is present (57). The most common polymorphism is C677T which causes a change from alanine to valine, rendering the enzyme thermolabile. Conflicting results have been reported regarding the prothrombotic role of the 677T variant (58). A meta-analysis suggested a weak effect (relative risk 1.2), but a recent population-based case-control study (the MEGA study) showed no association with VT (59). The prevalence varies widely according to populations, and is related to food content of folic acid. In Europe, there is a North to South gradient with a very high prevalence among Mediterranean countries. In addition, a certain level of microheterogeneity is seen in some populations, e.g., between northern and southern parts of Italy with allele frequencies of and 0.556, respectively (60). However, in Africans from the Sub- Saharan zone, this frequency is among the lowest in the world (38). The A1298C polymorphism occurs in the heterozygous state in approximately 9% 20% of most ethnic groups (61). By itself, MTHFR A1298C does not seem to be associated with hyperhomoysteinemia, but compound heterozygosity with MTHFR C677T results in decreased enzyme activity and increased homocysteine concentrations (62). MTHFR polymorphisms have also been associated with an increased risk of neural tube defects (63). It has been described that defects in the CBS gene may result in enzyme deficiency and hyperhomocysteinemia. A 68-bp insertion into the CBS gene (844ins68) has been described as a frequent mutation in various populations (64). Although this variant does not influence homocysteine concentrations or the risk of DVT (65); in combination with MTHFR C677T it may increase thrombotic risk (66). Other candidate genetic risk factors for VT O-blood group The moderate thrombotic risk associated with non-oo blood group (OR: ), and the high prevalence in the general population of prothrombotic non-oo genotypes ()50%), make the ABO blood group of general interest in thrombosis (67). The O blood group is associated with decreased concentrations of von Willebrand factor and factor VIII, whose increased plasma concentration represent an established risk factor for VTE (68). Miñano et al. studied the effect of ABO blood group on the risk and severity of VTs in carriers of FV Leiden and prothrombin 20210A polymorphisms. Their results confirmed the significant contribution of the non-oo blood group to the expression of VTE associated with FV Leiden wor: 1.76; (95% CI: )x. They suggested that the non-oo blood group also increases significantly the risk of VTE in carriers of prothrombin 20210A wor: 2.17; (95% CI: )x, in contrast with previous studies (69).

51 Bafunno and Margaglione: The role of genetic factors in venous thrombosis S45 Fibrinogen Abnormalities in fibrinogen have been reported to affect the risk for DVT. Qualitative deficiencies of fibrinogen (dysfibrinogenemias) have been found in patients with thrombosis and a prolonged thrombin time (70). Most of these patients have a mutation in the fibrinogen a-chain gene (FGA) or the g-chain gene (FGG). In the LETS, de Willige et al. investigated the association between haplotypes of FGA, FGB, and FGG genes, total fibrinogen concentrations, and the risk of DVT. The authors showed that the haplotype H2-tagging single nucleotide polymorphism (SNP) 10034C/T in the FGG gene reduced the fraction of the variant g9 chain in plasma, increasing the thrombotic risk (OR: 2.4; 95% CI, ) (71). Recently, the same authors evaluated whether FGG polymorphisms 10034C)T and 9340T)C were associated with risk of VTE in African Americans and Caucasians. In African Americans, 10034C)T and 9340T)C marginally influenced risk of VTE, with a 20% increase in risk for 10034TT carriers, and a 20% reduction in risk for 9340CC carriers. In Caucasians, 10034TT was associated with a 1.7-fold increase in risk, which increased to 2.1-fold for idiopathic VTE patients. 9340CC significantly reduced risk of VTE by approximately 2-fold (72). Although predisposition to thrombosis is a well-known feature of dysfibrinogenemia, relatively frequent thrombotic manifestations were seen in congenital afibrinogenemia. Simsek et al. reported a mutation analysis of a young afibrinogenemic man with multiple thrombo-embolic events involving both arteries and veins. Genetic analysis found a homozygous G to A transition in exon 5 of the Aa chain gene that predicted a homozygous W315X in the Aa chain. The patient was free of known risk factors as well as diseases associated with thrombosis. Therefore, it seems probable that afibrinogenemia itself might have contributed to both arterial and VT (73). FXIII The physiological role of FXIII in normal hemostasis is to improve the mechanical strength of the fibrin clot and to protect its elimination by the fibrinolytic system. In addition, it has also been implicated in the pathology of arterial and VT. Most recently, the polymorphisms in the FXIII subunit genes and their influence on the risk of thrombotic diseases have stirred much interest (74). Several studies have investigated the effect of FXIII Val34Leu polymorphism on the risk of DVT. There is wide variation in the prevalence of the Val34Leu mutation in different populations. In one study, the Val34Leu polymorphism was present in 2.5% of Asians, 28.9% of Blacks, 44.3% of Caucasians and 51.2% of American Indians. The 34Leu allele is virtually absent in Japanese, and present in 11% 27% of Australian Caucasians (75). Catto et al. determined the Val34Leu genotype in patients with DVT and PE, demonstrating a significant protection by the Leu34 allele wor: 0.63, (95% CI: )x (76). However, additional studies have not found a significant association. In 2006, Wells et al. published a comprehensive meta-analysis of 12 studies that showed a slight (OR: 0.89), but statistically significant protection afforded by the Leu34 allele (77). The Val34Leu polymorphism has been associated consistently with increased activation by thrombin in vitro. The clots formed in the presence of the FXIII 34Leu polymorphism are thinner and less porous. Thinner fibers are more resistant to fibrinolysis in vitro, and should theoretically increase the risk of a clinical thrombotic event. The putative protective effect of this SNP against VT is paradoxical. One study has suggested that the FXIII 34Leu polymorphism may protect against thrombosis, specifically in patients with high fibrinogen concentrations, who are known to be at increased risk for thrombotic complications (78). An interesting study was performed concerned the effect of a His95Arg substitution in the FXIII subunit B on the risk of DVT (74). This relationship was investigated in patients and controls from Leeds. The His/ArgqArg/Arg genotypes were found to be more frequent in patients than in controls (22.4% vs. 15.1%). FXIII-B His95Arg was also investigated in the LETS, in which a similar difference was observed between patients and controls (18.5% vs. 14.0%), with a pooled OR of 1.5 (CI: ) (74). Tissue factor pathway inhibitor (TFPI) TFPI is a circulating protease inhibitor that plays a major role in the inhibition of TF-factor VIIa proteolytic activity in vivo. Data obtained from the subjects enrolled in the LETS suggest that low concentrations of TFPI constitute a risk factor for DVT (79). The TFPI gene contains 9 exons that and have been mapped on 2q31-q32.1. To date, different polymorphisms have been described in the TFPI gene, but it is uncertain whether these polymorphisms influence the risk of thrombosis. A 399C)T transition in the 59 UTR of the gene was described, but this polymorphism neither alters the expression level of the TFPI protein, nor its association with the risk for VT (80). Moatti et al. performed an association study of 287 C)T in59utr, 33C)T transition in intron 7 with coronary artery disease and plasma TFPI concentrations. No association was found between these polymorphisms and coronary syndromes (81). No association between the Val264Met polymorphism and DVT has been described (82). However, a study of total TFPI antigen concentrations in 122 patients with DVT and 126 controls demonstrated an association between TFPI concentrations and VT (ps0.0001). However, these results do not support a link between DVT and Val264Met mutation. Kleesiek et al. described a Pro151Leu polymorphism in patients with a history of VT (83). They reported a statistically significant association between the C536T transition in the TFPI gene and the risk for VT (84). However, this result has not been confirmed in other populations (85). Thrombomodulin (TM) TM is a transmembrane protein expressed on the luminal surface of vascular endothelial cells. TM catalyzes thrombin activation of PC to APC, binds and alters thrombin substrate

52 S46 Bafunno and Margaglione: The role of genetic factors in venous thrombosis specificity, and catalyzes the inhibition of thrombin by antithrombin. Based on the important antithrombotic role of TM, it has hypothesized that polymorphisms within the TM gene that alter TM expression and/or anticoagulant function could predispose to VTE. Heit et al. screened a large Caucasian population containing unrelated patients with idiopathic DVT or PE to identify mutations within the TM gene and to assess the association with VTE. Four mutations, Ala25Thr, Ala455Val, 2470C deletion (39-UTR) and 4363A)G (39-flanking region) were more common in the study cohort, but were not associated with VTE by genotype or haplotype (86). However, Sugiyama et al. found that genetic variations in the TM gene, especially those with a haplotype consisting of 2729A)C SNP and A455V missense mutation, affect soluble TM concentrations and may be associated with DVT in the Japanese (87). Endothelial protein C receptor (EPCR) EPCR is a type I transmembrane protein expressed on the endothelium of large vessels that binds PC with high affinity. EPCR enhances the rate of PC activation by increasing the affinity of PC for the thrombin-thrombomodulin complex. In the last few years, several reports have suggested an association between mutations/polymorphisms in the EPCR gene and venous and arterial thrombosis (88). The human EPCR gene consists of 4 exons, spans approximately 6 kb and is located on chromosome 20q11.2. Several variations have been reported in the EPCR gene, some of which are associated with the risk of thrombosis. The first EPCR gene mutation to be described was a 23 bp insertion in exon 3 resulting in a stop codon and truncated protein. Given its low frequency in the population (-1%), it will be difficult to assess the effect of this mutation on the risk of venous and arterial thrombosis (89). Moreover, different point mutations have been reported in the EPCR gene promoter, but they were rare in patients with VTE or myocardial infarction. A substitution of Arg for Cys at position 98 was identified with a prevalence of 0.44% in patients and 0.83% in controls, suggesting no role for this mutation in VTE. Four haplotypes have been reported in the EPCR gene; H1, H2, H3 and H4. Three haplotypes contain one or more SNPs that are haplotype-specific. The H3 haplotype, tagged by the rare allele of 4600A/G, is associated with increased plasma concentrations of sepcr, but its association with risk of VTE is controversial. The H1 haplotype, tagged by the rare allele of 4678G/C, was reported to be associated with a decreased risk of VTE wors0.59, (95% CI, )x as well as in carriers of FV Leiden. Moreover, APC concentrations were found to be significantly higher in carriers of the H1H1 genotype than in those not having the H1 haplotype, both in patients and in controls and resulting in a protective effect against VTE. It has been observed that individuals carrying the H3 haplotype present with increased sepcr concentrations that could increase the risk of VT. Finally, the H4 haplotype was reported to be associated with a slight increase in the risk of VTE (90). Thrombin activatable fibrinolysis inhibitor (TAFI) TAFI is an important inhibitor of fibrinolysis which acts by inhibiting the assembly of fibrinolytic factors on the fibrin surface. It has been shown that increased TAFI antigen concentrations are associated with a mild (1.7-fold) but significantly increased risk for DVT (91). The human TAFI gene has been mapped to chromosome 13q14.11, spans approximately 48 kb and is organised in 11 exons (92). TAFI concentrations are also determined genetically, and the 438 G/A SNP in the promoter region and the 505 G/A SNP and 1040 C/T SNP in the coding region are associated with TAFI plasma antigen concentrations. The 438 G/A polymorphism was reported to be associated with an increased risk for development of VT (93). Martini et al. reported that the 505 G-allele is associated with low TAFI antigen concentrations and carriers of the 505G allele showed a mildly increased risk of VT (94). P-selectin P-selectin, a member of the cell adhesion molecules, is primarily stored in the a-granules of platelets and the Weibel- Palade bodies of endothelial cells. The interaction of P-selectin and its main-counter receptor leads to several mechanisms that induce a procoagulant state by triggering the generation of procoagulant microparticles from leukocytes, upregulating the expression of tissue factor on monocytes and inducing an increased surface-dependent thrombin generation on monocytes. High plasma concentrations of soluble P-selectin (sp-selectin) are strongly associated with VTE (95). Moreover, sp-selectin concentrations were found to be higher in cancer patients with VTE compared to those without VTE, allowing prediction of the occurrence of cancer-associated VTE, with a hazard ratio of 2.6 (95% CI: ) (96). A number of P-selectin gene variants that affect the protein sequence have been described. One such variant, Thr715Pro, has consistently been associated with lower spselectin concentrations in plasma. Ay et al. investigated spselectin and the Thr715Pro variant in a high-risk population of patients with a history of recurrent VTE. This study demonstrated that sp-selectin concentrations are lower in individuals carrying the Pro715 variant than those without. Therefore, the Pro715 variant may be seen as a protective allele, providing defense against increases in sp-selectin concentration that are commonly seen in Thr715 homozygotes (97). Genome wide association studies During the last years, several investigations have been performed where a great number of SNPs are tested in association studies in order to fully understand the genetic causes of thrombosis. Bezemer and colleagues investigated almost 20,000 potentially functional SNPs for association with DVT. They used three case-control studies derived from the LETS and the

53 Bafunno and Margaglione: The role of genetic factors in venous thrombosis S47 MEGA study. The evidence for an association with DVT was strongest for the three SNPs (p-0.05): rs in CYP4V2 (risk allele frequency, 0.64), rs in SER- PINC1 (risk allele frequency, 0.10), and rs in GP6 (risk allele frequency, 0.84). The OR for DVT per risk allele was 1.24 (95% CI, ) for rs , 1.29 (95% CI, ) for rs , and 1.15 (95% CI, ) for rs It is interesting to note that these SNPs are in or near genes that have a clear role in blood coagulation. In the region of CYP4V2, there have been identified four additional SNPs in the CYP4V2/KLKB1/F11 locus that were also associated with both DVT (highest OR per risk allele, 1.39; 95% CI, ) and coagulation factor XI concentrations (highest increase per risk allele, 8%; 95% CI, 5% 11%) (98). One of the SNPs identified was an A)G sequence variant in the gene encoding factor IX (rs6048, also known as F9 Malmö). The G allele was associated with a 15% 43% decrease in risk of DVT compared with the A allele. This common variant (minor allele frequencys0.32) leads to the substitution of alanine for threonine at amino acid 148 in the portion that is cleaved to activate factor IX. The OR for the G allele of F9 Malmö, compared with the A allele, was 0.80 (95% CI, ). The SNP rs in FIX was strongly linked to F9 Malmö, and was similarly associated with DVT. No other SNP or haplotype tested was more strongly associated. However, the biological mechanism by which F9 Malmö influences the risk of DVT remains unclear (99). It has been suggested that the overall effect of the major proinflammatory cytokine interleukin-1 (IL-1) on coagulation and fibrinolysis is prothrombotic. Van Minkelen et al. have investigated whether common variations in the genes coding for IL-1b, IL-1Ra, IL-1R1, and IL-1R2 (IL1B, IL1RN, IL1R1, and IL1R2) influence the risk of VT by modulating the IL-1 pathway. The authors genotyped 18 SNPs in these genes, together with the Tag 25 haplotype groups, in all patients and control subjects in the LETS. Global testing using a recessive model showed a difference in haplotype frequency between patients and controls for IL1RN (ps0.031). Subsequently, the risk of VT was calculated for each haplotype of IL1RN. Increased thrombotic risk was found for homozygous carriers of haplotype 5 (H5, tagged by SNP 13888T/G, rs ) of IL1RN (ORs3.9; 95% CI: ; ps0.002) (100). Recently, Reiner et al. screened 290 common SNPs in 51 coagulation-fibrinolysis and inflammation candidate genes for association with VTE in the Cardiovascular Health Study (CHS), a large, well-characterized prospective populationbased study of older adults. TagSNPs within four genes encoding factor XIII subunit A (F13A), factor VII activating protease (HABP2), protease activated receptor-1 (F2R), and the urokinase receptor (PLAUR) showed the strongest evidence for association with VTE. The rs variant allele of F13A1 was associated with a 1.66-fold increased risk of VTE, while the minor alleles of HABP2 rs and rs , F2R rs and rs153311, and PLAUR rs were each associated with lower risk of VTE (hazard ratios in the range of ). Consistent with the observed protective association on VTE risk, the HABP2 rs variant allele was also associated with lower activity levels of coagulation factors VIII, IX, X, and plasminogen (101). Conclusions VT represents a significant public health problem in Western countries resulting from the interaction of genetic predisposition and environmental risk factors. In this review we have focused on the role of genetic risk factors, mainly related to the haemostatic system, in influencing thrombotic risk. The most significant evidence emerging from genetic studies over the past few decades has been the confirmation that inherited hypercoagulable conditions are present in a large proportion of patients with venous thromboembolic disease. These include mutations in the genes that encode AT, PC and PS, that may be viewed as strong risk factors, and the FV Leiden and factor II G20210 A mutations commonly referred to as milder risk factors. The gain-of-function mutations in procoagulant factors were more prevalent than abnormalities in anticoagulant proteins in the typical Caucasian population. Thus, the focus shifted from family studies to populationbased association studies in a case-control format. In fact, common genetic variations in coagulation proteins were recently tested in association studies. In addition to these well-established conditions predisposing to VT, different studies have pointed to the role of other candidate genetic risk factors, such as variants in FXIII, FIX and fibrinogen genes. Moreover, less-well studied proteins that are part of anticoagulant pathways, such as TFPI, TM and EPCR, have also been investigated as risk factors for VT. Recently, different genome-wide association studies have been performed where several SNPs are tested in association studies and linked to thrombotic risk. This extensive list of genetic factors illustrates that VT can be understood by assuming an interaction between different mutations in candidate susceptible genes. By itself, the risk associated with each genetic defect may be relatively low, but the simultaneous presence of several mutations may dramatically increase disease susceptibility. Moreover, environmental factors may interact with one or more genetic variations to add further to the risk. Even considering the occurrence of gene-gene interaction, it is interesting to note that the known thrombophilic abnormalities do not explain all cases of variable penetrance of VT in families with a certain identified defect. This evidence points to the existence of additional unidentified prothrombotic defects that remain to be discovered. Conflict of interest statement Authors conflict of interest disclosure: The authors stated that there are no conflicts of interest regarding the publication of this article. Research funding: None declared.

54 S48 Bafunno and Margaglione: The role of genetic factors in venous thrombosis Employment or leadership: None declared. Honorarium: None declared. References 1. Macfarlane RG. Haemostasis: introduction. Br Med Bull 1977;33: Reistma PH, Rosendaal FR. Past and future of genetic research in thrombosis. J Thromb Haemost 2007;5: Naess IA, Christiansen SC, Romundstad P, Cannegieter SC, Rosendaal FR, Hammerstrøm J. Incidence and mortality of venous thrombosis: a population-based study. J Thromb Haemost 2007;5: Kyrle PA, Minar E, Bialonczyk C, Hirschl M, Weltermann A, Eichinger S. The risk of recurrent venous thromboembolism in men and women. N Engl J Med 2004;350: Bronić A. Thromboembolic diseases as biological and clinical syndrome role of the Mediterranean League against Thromboembolic Diseases. Biochem Med 2010;20: Souto JC, Almasy L, Borrell M, Blanco-Vaca F, Mateo J, Soria JM, et al. Genetic susceptibility to thrombosis and its relationship to physiological risk factors: the GAIT study. Am J Hum Genet 2000;67: Briggs J. Recurrent phlebitis of obscure origin Bull. John Hopkins Hosp 1905;16: Rosendaal FR, Reitsma PH. Genetics of venous thrombosis. J Thromb Haemost 2009;7: Bezemer ID, van der Meer FJ, Eikenboom JC, Rosendaal FR, Doggen CJ. The value of family history as a risk indicator for venous thrombosis. Arch Intern Med 2009;169: Tait RC, Walker ID, Perry DJ, Islam SI, Daly ME, McCall F, et al. Prevalence of antithrombin deficiency in the healthy population. Br J Haematol 1994;87: van Boven HH, Lane DA. Antithrombin and its inherited deficiency states. Semin Hematol 1997;34: Lane DA, Mannucci PM, Bauer KA, Bertina RM, Bochkov NP, Boulyjenkov V, et al. Inherited thrombophilia: Part I. Thromb Haemost 1996;76: Lane DA, Bayston T, Olds RJ, Fitches AC, Cooper DN, Millar DS, et al. Antithrombin mutation database: 2nd (1997) update. For the Plasma Coagulation Inhibitors Subcommittee of the Scientific and Standardization Committee of the International Society on Thrombosis and Haemostasis. Thromb Haemost 1997;77: Olds RJ, Lane DA, Chowdhury V, De Stefano V, Leone G, Thein SL. Complete nucleotide sequence of the antithrombin gene: evidence for homologous recombination causing thrombophilia. Biochemistry 1993;32: van Boven HH, Olds RJ, Thein SL, Reistma PH, Lane DA, Briet E, et al. Hereditary antithrombin deficiency: heterogeneity of the molecular basis and mortality in Dutch families. Blood 1994;84: Patnaik MM, Moll S. Inherited antithrombin deficiency: a review. Haemophilia 2008;14: Demers C, Ginsberg JS, Hirsh J, Henderson P, Blajchman MA. Thrombosis in antithrombin III-deficient persons: report of a large kindred and literature review. Ann Intern Med 1992;116: Bucciarelli P, Rosendaal FR, Tripodi A, Mannucci PM, De Stefano V, Palareti G, et al. Risk of venous thromboembolism and clinical manifestations in carriers of antithrombin, protein C, protein S deficiency, or activated protein C resistance. A multicenter collaborative family study. Arterioscler Thromb Vasc Biol 1999;19: Broekmans AW, Veltkamp JJ, Bertina RM. Congenital protein C deficiency and venous thromboembolism: a study of three Dutch families. N Engl J Med 1983;309: Schwarz HP, Fischer M, Hopmeier P, Batard MA, Griffin JH. Plasma protein S deficiency in familial thrombotic disease. Blood 1984;64: Allaart CF, Poort SR, Rosendaal FR, Reitsma PH, Bertina RM, Briet E. Increased risk of venous thrombosis in carriers of hereditary protein C deficiency defect. Lancet 1993;341: Reitsma PH. Protein C deficiency: from gene defects to disease. Thromb Haemost 1997;8: Millar DS, Johansen B, Berntorp E, Minford A, Bolton-Maggs P, Wensley R, et al. Molecular genetic analysis of severe protein C deficiency. Hum Genet 2000;106: Reitsma PH, Bernardi F, Doig RG, Gandrille S, Greengard JS, Ireland H, et al. Protein C deficiency: a database of mutations, 1995 update. On behalf of the Subcommittee on Plasma Coagulation Inhibitors of the Scientific and Standardization Committee of the ISTH. Thromb Haemost 1995;73: Makris M, Leach M, Beauchamp NJ, Daly ME, Cooper PC, Hampton KK, et al. Genetic analysis, phenotypic diagnosis, and risk of venous thrombosis in families with inherited deficiencies of protein S. Blood 2000;95: García de Frutos P, Fuentes-Prior P, Hurtado B, Sala N. Molecular basis of protein S deficiency. Thromb Haemost 2007;98: Pintao MC, Garcia AA, Borgel D, Alhenc-Gelas M, Spek CA, de Visser MC, et al. Gross deletions/duplications in PROS1 are relatively common in point mutation-negative hereditary protein S deficiency. Hum Genet 2009;126: Koster T, Rosendaal FR, Briët E, van der Meer FJ, Colly LP, Trienekens PH, et al. Protein C deficiency in a controlled series of unselected outpatients: an infrequent but clear risk factor for venous thrombosis (Leiden Thrombophilia Study). Blood 1995;85: Vossen CY, Conard J, Fontcuberta J, Makris M, van der Meer FJ, Pabinger I, et al. Familial thrombophilia and lifetime risk of venous thrombosis. J Thromb Haemost 2004;2: Heijboer H, Brandjes DP, Büller HR, Sturk A, Ten Cate JW. Deficiencies of coagulation inhibiting and fibrinolytic proteins in outpatients with deep-vein thrombosis. N Engl J Med 1990;323: Zöller B, Berntsdotter A, Garcia de Frutos P, Dahlback B. Resistance to activated protein C as an additional genetic risk factor in hereditary deficiency of protein S. Blood 1995; 85: Hasstedt SJ, Bovill EG, Callas PW, Long GL. An unknown genetic defect increases venous thrombosis risk, through interaction with protein C deficiency. Am J Hum Genet 1998; 63: Mahmoodi BK, Brouwer JL, Veeger NJ van der Meer J. Hereditary deficiency of protein C or protein S confers increased risk of arterial thromboembolic events at a young age: results from a large family cohort study. Circulation 2008;118: Dahlbäck B, Carlsson M, Svensson P. Familial thrombophilia due to a previously unrecognized mechanism characterized by poor anticoagulant response to activated protein C: prediction of a cofactor to activated protein C. Proc Natl Acad Sci USA 1993;90:

55 Bafunno and Margaglione: The role of genetic factors in venous thrombosis S Bertina RM, Koeleman RP, Koster T, Rosendaal FR, Dirven RJ, de Ronde H, et al. Mutation in blood coagulation factor V associated with resistance to activated protein C. Nature 1994;369: Martinelli I, Mannucci PM, De Stefano V, Taioli E, Rossi V, Crosti F, et al. Different risks of thrombosis in four coagulation defects associated with inherited thrombophilia: a study of 150 families. Blood 1998;92: Rosendaal FR, Koster T, Vandenbroucke JP, Reitsma PH. High risk of thrombosis in patients homozygous for factor V Leiden (activated protein C resistance). Blood 1995;85: Bauduer F, Lacombe D. Factor V Leiden, prothrombin 20210A, methylenetetrahydrofolate reductase 677T, and population genetics. Mol Genet Metab 2005;86: Rees DC, Cox M, Clegg JB. World distribution of factor V Leiden. Lancet 1995;346: Griffin JH, Xu X, Pabinger I, Samama M, Conard J, Brenner B, et al. A single genetic origin for a common Caucasian risk factor for venous thrombosis. Blood 1997;89: Lindqvist PG, Svensson PJ, Dahlback B, Marsal K. Factor V Q506 mutation (activated protein C resistance) associated with reduced intrapartum blood loss a possible evolutionary selection mechanism. Thromb Haemost 1998;79: Alfirevic Z, Simundic AM, Nikolac N, Sobocan N, Alfirevic I, Stefanovic M, et al. Frequency of factor II G20210A, factor V Leiden, MTHFR C677T and PAI-1 5G/4G polymorphism in patients with venous thromboembolism: croatian case-control study. Biochem Med 2010;20: De Stefano V, Martinelli I, Mannucci PM, Paciaroni K, Chiusolo P, Casorelli I, et al. The risk of recurrent deep venous thrombosis among heterozygous carriers of both factor V Leiden and the G20210 A prothrombin mutation. N Engl J Med 2000;341: Margaglione M, D Andrea G, Colaizzo D, Cappucci G, del Popolo A, Brancaccio V, et al. Coexistence of factor V Leiden and factor II A20210 mutation and recurrent venous thromboembolism. Thromb Haemost 1999;82: Simioni P, Prandoni P, Lensing AW, Scudeller A, Sardella C, Prins MH, et al. The risk of recurrent venous thromboembolism in patients with an Arg506 Gln mutation in the gene for factor V. N Engl J Med 1997;336: Bernardi F, Faioni EM, Castoldi E, Lunghi B, Castaman G, Sacchi E, et al. A factor V genetic component differing from factor V R506Q contributes to the activated protein C resistance phenotype. Blood 1997;90: Williamsson D, Brown K, Luddington R, Baglin C, Baglin T. A new mutation (Arg306Thr) associated with resistance to activated protein C. Blood 1998;91: Chan WP, Lee CK, Kwong YL, Lam CK, Liang R. A novel mutation of Arg306 of factor V gene in Hong Kong Chinese. Blood 1998;91: Poort SW, Rosendaal FR, Reitsma PH, Bertina RM. A common genetic variation in the 3 -untranslated region of the prothrombin gene is associated with elevated plasma prothrombin levels and an increase in venous thrombosis. Blood 1996; 88: Rosendaal FR, Doggen CJ, Zivelin A, Arruda VR, Aiach M, Siscovick DS, et al. Geographic distribution of the G to A prothrombin variant. Thromb Haemost 1998;79: Lijfering WM, Middeldorp S, Veeger JG, Hamulyak K, Prins MH, Buller HR, et al. Risk of recurrent venous thrombosis in homozygous carriers and double heterozygous carriers of factor V Leiden and prothrombin G20210A. Circulation 2010; 121: Simioni P, Prandoni P, Burlina A, Tormene D, Sardella C, Ferrari V, et al. Hyperhomocysteinemia and deep vein thrombosis: a case control study. Thromb Haemost 1996;76: Mudd SH, Skovby F, Levy HL, Pettigrew KD, Wilcken B, Pyeritz RE, et al. The natural history of homocystinuria due to cystathionine beta-synthase deficiency. Am J Hum Genet 1985;37: Goyette P, Frosst P, Rosenblatt DS, Rozen R. Seven novel mutations in the methylenetetrahydrofolate reductase gene and genotype/phenotype correlations in severe methylenetetrahydrofolate reductase deficiency. Am J Hum Genet 1995;56: Sibani S, Christensen B, O Ferrall E, Saadi I, Hiou-Tim F, Rosenblatt DS, et al. Characterization of six novel mutations in the methylenetetrahydrofolate reductase (MTHFR) gene in patients with homocystinuria. Hum Mutat 2000;15: Kraus JP, Janosík M, Kozich V, Mandell R, Shih V, Sperandeo MP, et al. Cystathionine beta-synthase mutations in homocystinuria. Hum Mutat 1999;13: Pereira AC, Schettert IT, Morandini Filho AA, Guerra-Shinohara EM, Krieger JE. Methylenetetrahydrofolate reductase (MTHFR) C677T gene variant modulates the homocysteine folate correlation in a mild folate-deficient population. Clin Chim Acta 2004;340: Tsai AW, Cushman M, Tsai MY, Heckbert SR, Rosamond WD, Aleksic N, et al. Serum homocysteine, thermolabile variant of methylenetetrahydrofolate reductase (MTHFR) and venous thromboembolism: longitudinal Investigation of thromboembolism Etiology (LITE). Am J Hematol 2003;72: Bezemer ID, Doggen CJ, Vos HL, Rosendaal FR. No association between the common MTHFR 677C)T polymorphism and venous thrombosis: results from MEGA study. Arch Intern Med 2007;167: Pepe G, Camacho Vanegas O, Giusti B, Brunelli T, Marcucci R, Attanasio M, et al. Heterogeneity in world distribution of the thermolabile C677T mutation in 5,10-methylenetetrahydrofolate reductase. Am J Hum Genet 1998;63: Esfahani ST, Cogger EA, Caudill MA. Heterogeneity in the prevalence of methylenetetrahydrofolate reductase gene polymorphisms in women of different ethnic groups. J Am Diet Assoc 2003;103: Weisberg I, Tran P, Christensen B, Sibani S, Rozen R. A second genetic polymorphism in methylenetetrahydrofolate reductase (MTHFR) associated with decreased enzyme activity. Mol Genet Metab 1999;68: Botto LD, Yang Q. 5,10-Methylenetetrahydrofolate reductase gene variants and congenital anomalies: a HuGE review. Am J Epidemiol 2000;151: Franco RF, Elion J, Lavinha J, Krishnamoorthy R, Tavella MH, Zago MA. Heterogeneous ethnic distribution of the 844ins68 in the cystathionine b-synthase gene. Hum Hered 1998;48: Franco R, Maffei F, Lourenco D, Piccinato C, Morelli V, Thomazini I, et al. The frequency of 844ins68 mutation in the cystathionine b-synthase gene is not increased in patients with venous thrombosis. Haematologica 1998;83: de Franchis R, Fermo I, Mazzola G, Sebastio G, Di Minno G, Coppola A, et al. Contribution of the cystathionine betasynthase gene (844ins68) polymorphism to the risk of earlyonset venous and arterial occlusive disease and fasting hyperhomocysteinemia. Thromb Haemost 2000;84:

56 S50 Bafunno and Margaglione: The role of genetic factors in venous thrombosis 67. Jick H, Slone D, Westerholm B, Inman WH, Vessey MP, Shapiro S, et al. Venous thromboembolic disease and ABO blood type. A cooperative study. Lancet 1969;1: Morelli VM, De Visser MC, Vos HL, Bertina RM, Rosendaal FR. ABO blood group genotypes and the risk of venous thrombosis: effect of factor V Leiden. J Thromb Haemost 2005;3: Miñano A, Ordóñez A, España F, González-Porras JR, Lecumberri R, Fontcuberta J, et al. ABO blood group and risk of venous or arterial thrombosis in carriers of factor V Leiden or prothrombin G20210A polymorphisms. Haematologica 2008;93: Mosesson MW. Dysfibrinogenemia and thrombosis. Semin Thromb Hemost 1999;25: de Willige SU, de Visser MC, Houwing-Duistermaat JJ, Rosendaal FR, Vos HL, Bertina RM. Genetic variation in the fibrinogen gamma gene increases the risk for deep venous thrombosis by reducing plasma fibrinogen g levels. Blood 2005;106: de Willige SU, Pyle ME, Vos HL, de Visser MC, Lally C, Dowling NF, et al. Fibrinogen gamma gene 3 -end polymorphisms and risk of venous thromboembolism in the African- American and Caucasian population. Thromb Haemost 2009; 101: Simsek I, de Mazancourt P, Horellou MH, Erdem H, Pay S, Dinc A, et al. Afibrinogenemia resulting from homozygous nonsense mutation in A alpha chain gene associated with multiple thrombotic episodes. Blood Coagul Fibrinolysis 2008; 19: Muszbek L, Bagoly Z, Bereczky Z, Katona E. The involvement of blood coagulation factor XIII in fibrinolysis and thrombosis. Cardiovasc Hematol Agents Med Chem 2008; 6: Attié-Castro FA, Zago MA, Lavinha J, Elion J, Rodriguez- Delfin L, Guerreiro JF, et al. Ethnic heterogeneity of the factor XIII Val34Leu polymorphism. Thromb Haemost 2000;84: Catto AJ, Kohler HP, Coore J, Mansfield MW, Stickland MH, Grant PJ. Association of a common polymorphism in the factor XIII gene with venous thrombosis. Blood 1999;93: Wells PS, Anderson JL, Scarvelis DK, Doucette SP, Gagnon F. Factor XIII Val34Leu variant is protective against venous thromboembolism: a HuGE review and meta-analysis. Am J Epidemiol 2006;164: Vossen CY, Rosendaal RF. Protective effect of the factor XIII/ Val34Leu mutation on the risk of deep venous thrombosis is dependent on the fibrinogen level. J Thromb Haemost 2005;3: Dahm A, van Hylckama Vlieg A, Bendz B, Rosendaal F, Bertina RM, Sandset PM. Low levels of tissue factor pathway inhibitor (TFPI) increase the risk of venous thrombosis. Blood 2003;101: Miyata T, Sakata T, Kumeda K, Uchida K, Tsushima M, Fujimura H, et al. C-399T polymorphism in the promoter region of human tissue factor pathway inhibitor TFPI gene does not change the plasma TFPI antigen level and does not cause venous thrombosis. Thromb Haemost 1998;80: Moatti D, Haidar B, Fumeron F, Gauci L, Boudvillain O, Seknadji P, et al. A new T-287 C polymorphism in the 5 regulatory region of the tissue factor pathway inhibitor gene. Association study of the T-287 C and C-399 T polymorphisms with coronary artery disease and plasma TFPI levels. Thromb Haemost 2000;84: Amini-Nekoo A, Futers TS, Moia M, Mannucci PM, Grant PJ, Ariëns RA. Analysis of the tissue factor pathway inhibitor gene and antigen levels in relation to venous thrombosis. Br J Haematol 2000;113: Kleesiek K, Schmidt M, Götting C, Brinkmann T, Prohaska W. A first mutation in the human tissue factor pathway inhibitor gene encoding wp151lxtfpi. Blood 1998;92: Kleesiek K, Schmidt M, Gotting C, Schwenz B, Lange S, Muller-Berghaus G, et al. The 536C T transition in the human tissue factor pathway inhibitor (TFPI) gene is statistically associated with a higher risk for venous thrombosis. Thromb Haemost 1999;82: Gonzalez-Conejero R, Lozano ML, Corral J, Martinez C, Vicente V. The TFPI 536C T mutation is not associated with increased risk for venous or arterial thrombosis. Thromb Haemost 2000;83: Heit JA, Petterson TM, Owen WG, Burke JP, de Andrade M, Melton J. Thrombomodulin gene polymorphisms or haplotypes as potential risk factors for venous thromboembolism: a population-based case control study. J Thromb Haemost 2004;3: Sugiyama S, Hirota H, Kimura R, Kokubo Y, Kawasaki T, Suehisa E, et al. Haplotype of thrombomodulin gene associated with plasma thrombomodulin level and deep vein thrombosis in the Japanese population. Thromb Res 2007;119: Medina P, Navarro S, Estellés A, España F. Polymorphisms in the endothelial protein C receptor gene and thrombophilia. Thromb Haemost 2007;98: Poort SR, Vos HL, Rosendaal FR, Bertina RM. The endothelial protein C receptor (EPCR) 23 bp insert mutation and the risk of venous thrombosis. Thromb Haemost 2002;88: Medina P, Navarro S, Estellés A, Vayà A, Woodhams B, Mira Y, et al. Contribution of polymorphisms in the endothelial protein C receptor gene to soluble endothelial protein C receptor and circulating activated protein C levels, and thrombotic risk. Thromb Haemost 2004;91: van Tilburg NH, Rosendaal FR, Bertina RM. Thrombin activatable fibrinolysis inhibitor and the risk for deep vein thrombosis. Blood 2000;95: Boffa MB, Reid ST, Joo E, Nesheim ME, Koschinsky ML. Characterization of the gene encoding human TAFI (thrombinactivatable fibrinolysis inhibitor; plasma carboxypeptidase B). Biochemistry 1999;38: Franco RF, Fagundes MG, Meijers JC, Reitsma PH, Lourenco D, Morelli V, et al. Identification of polymorphisms in the 5 - untranslated region of the TAFI gene: relationship with plasma TAFI levels and risk of venous thrombosis. Haematologica 2001;86: Martini CH, Brandts A, de Bruijne EL, van Hylckama Vlieg A, Leebeek FW, Lisman T, et al. The effect of genetic variants in the thrombin activatable fibrinolysis inhibitor (TAFI) gene on TAFI-antigen levels, clot lysis time and the risk of venous thrombosis. Br J Haematol 2006;134: Pabinger I, Ay C. Biomarkers and venous thromboembolism. Arterioscler Thromb Vasc Biol 2009;29: Ay C, Simanek R, Vormittag R, Dunkler D, Alguel G, Koder S, et al. High plasma levels of soluble P-selectin are predictive of venous thromboembolism in cancer patients: results from the Vienna Cancer and Thrombosis Study (CATS). Blood 2008;112: Ay C, Jungbauer LV, Sailer T, et al. High concentrations of soluble P-selectin are associated with risk of venous throm-

57 Bafunno and Margaglione: The role of genetic factors in venous thrombosis S51 boembolism and the P-selectin Thr715 variant. Clin Chem 2007;53: Bezemer ID, Bare LA, Doggen CJ, Arellano AR, Tong C, Rowland CM, et al. Gene variants associated with deep vein thrombosis. J Am Med Assoc 2008;299: Bezemer ID, Arellano AR, Tong CH, Rowland CM, Ireland HA, Bauer KA, et al. F9 Malmö, factor IX and deep vein thrombosis. Haematologica 2009;94: van Minkelen R, de Visser MC, Houwing-Duistermaat JJ, Vos HL, Bertina RM, Rosendaal FR. Haplotypes of IL1B, IL1RN, IL1R1, and IL1R2 and the risk of venous thrombosis. Arterioscler Thromb Vasc Biol 2007;27: Reiner AP, Lange LA, Smith NL, Zakai NA, Cushman M, Folsom AR. Common hemostasis and inflammation gene variants and venous thrombosis in older adults from the Cardivascular Health Study. J Thromb Haemost 2009;7:

58

59 Clin Chem Lab Med 2010;48(Suppl 1):S53 S by Walter de Gruyter Berlin New York. DOI /CCLM Protein C and protein S deficiencies: similarities and differences between two brothers playing in the same game Zsuzsanna Bereczky*, Kitti B. Kovács and László Muszbek Clinical Research Center and Thrombosis, Haemostasis and Vascular Biology Research Group of the Hungarian Academy of Sciences, University of Debrecen, Medical and Health Science Center, Debrecen, Hungary Abstract Protein C (PC) and protein S (PS) are vitamin K-dependent glycoproteins that play an important role in the regulation of blood coagulation as natural anticoagulants. PC is activated by thrombin and the resulting activated PC (APC) inactivates membrane-bound activated factor VIII and factor V. The free form of PS is an important cofactor of APC. Deficiencies in these proteins lead to an increased risk of venous thromboembolism; a few reports have also associated these deficiencies with arterial diseases. The degree of risk and the prevalence of PC and PS deficiency among patients with thrombosis and in those in the general population have been examined by several population studies with conflicting results, primarily due to methodological variability. The molecular genetic background of PC and PS deficiencies is heterogeneous. Most of the mutations cause type I deficiency (quantitative disorder). Type II deficiency (dysfunctional molecule) is diagnosed in approximately 5% 15% of cases. The diagnosis of PC and PS deficiencies is challenging; functional tests are influenced by several pre-analytical and analytical factors, and the diagnosis using molecular genetics also has special difficulties. Large gene segment deletions often remain undetected by DNA sequencing methods. The presence of the PS pseudogene makes genetic diagnosis even more complicated. Clin Chem Lab Med 2010;48:S Keywords: protein C deficiency; protein S deficiency; review; thrombophilia. Introduction Venous thromboembolism (deep vein thrombosis and/or pulmonary embolism, VTE) and its consequences occur with high frequency Western societies. VTE is still a major cause *Corresponding author: Zsuzsanna Bereczky, MD, PhD, Clinical Research Center, University of Debrecen, Medical and Health Science Center, 98, Nagyerdei krt., Debrecen, 4032, Hungary Phone: q , Fax: q , zsbereczky@med.unideb.hu Received July 4, 2010; accepted September 26, 2010; previously published online November 6, 2010 of morbidity and mortality during pregnancy and stillbirth, and occurs with a relatively high frequency in young women using oral contraceptives. VTE can be recurrent and may also lead to post-thrombotic syndrome, a chronic disease with disabling pain and ulceration. VTE is a typical example of common complex diseases; both acquired (environmental) and genetic causes play an important role in the development of the disease (1). Over the last decades, several genetic risk factors for VTE have been identified. Loss of function mutations in different components of the natural anticoagulant system lead to antithrombin III (ATIII), protein C (PC) and protein S (PS) deficiencies, while gain of function mutations in the genes of coagulation factors, such as Factor V Leiden (FVL) and prothrombin 20210A allele are responsible for the majority of inherited thrombophilia. The aim of this review is to give an overview of the physiology of PC and PS, on the molecular basis of their deficiencies and on the laboratory diagnosis of these disorders including the difficulties and challenges in this field. The role of protein C-protein S system in the regulation of coagulation PC and PS play important roles in the regulation of blood coagulation as natural anticoagulants (2, 3). PC is activated by thrombin in the presence of thrombomodulin (TM). TM is an endothelial cell surface protein, and upon binding, thrombin becomes a potent activator of PC. Endothelial protein C receptor (EPCR) also is highly important in the activation process; EPCR binds PC through its Gla-domain and presents it to the thrombin-tm complex. Thrombin cleaves the activation peptide domain of PC at Arg169 resulting in a 12-amino acid long activation peptide being released from the N-terminal end of its heavy chain. Activated PC (APC) inactivates membrane-bound active factor VIII (FVIIIa) and factor V (FVa) by cleaving these factors specific arginine residues. FVa is cleaved at Arg506, which is the preferred cleavage site. However, full inactivation of FVa also requires cleavage at Arg306. Cleavage at Arg679 seems unimportant for inactivation of FVa. FVIIIa is cleaved at Arg336 and Arg562. Non-activated forms of FVIII and FV are poor substrates for APC. Esmon and co-workers showed that the membrane phospholipid requirement for the anticoagulant APC complex differs from that of the procoagulant complexes (4). Phosphatidyl ethanolamine, instead of phosphatidyl serine, is required for the binding of the APC complex to the membrane of endothelial cells. APC can also cleave intact FV at Arg506, which makes FV a cofactor of APC in inactivating FVIIIa. The main inhibitor of APC is protein C 2010/392

60 S54 Bereczky et al.: Protein C and protein S deficiencies inhibitor, a single chain glycoprotein serine protease inhibitor synthesized in the liver. The inhibitor forms a 1:1 complex with APC and is cleaved at the reactive site (Arg354). APC is also inhibited by a-1 antitrypsin. In addition to its anticoagulant function, PC plays an important role in cytoprotection, which has been partially clarified in the last decade. The multiple cytoprotective effects of APC is not the subject of this review, it has been reviewed by Mosnier et al. (5). The free form of PS (described later) is an important cofactor of APC, enhancing its affinity to negative charged phospholipid surfaces. PS is able to displace FXa from its complex with FVa, allowing APC to cleave FVa at Arg506. In the process of FVIIIa inactivation APC activity is synergistically stimulated by PS and the non-activated form of FV, while in the process of FVa inactivation, free PS is the only cofactor of APC. PS also forms a complex with the complement 4b binding protein (C4bBP); this complex lacks APC cofactor activity. PS also has direct, APC-independent effects (6, 7). PS binds to FVa present on phospholipid vesicles, and inhibits prothrombinase activity by competing with prothrombin for binding to FVa. Binding of PS to FXa has also been demonstrated. APC-independent PS function does not seem to be restricted to the free form; C4bBP-complexed PS has the same activity. A direct interaction of PS with tissue factor pathway inhibitor (TFPI) in the inhibition of FXa was recently reported (8). Protein and gene structure of protein C PC is a vitamin K-dependent glycoprotein synthesized by the liver as a single chain protein. It exists in the plasma as a precursor of a serine protease at a concentration of 3 5 mg/l (9). Its half-life is short, approximately 8 h in the circulation. The mature 62 kda protein is composed of a heavy (41 kda) and a light (21 kda) chain, these chains are held together by a single disulfide bond between Cys141 and Cys265. The domain structure of PC shows high similarity to other vitamin K-dependent proteins. It has a pre-pro leader sequence (numbered as 42 to 1 by the traditional numbering system, where the first methionine corresponds to 42), which is required for g-carboxylation of glutamic acid residues in the Gla-domain and for secretion. The mature protein contains a Gla-domain (amino acids 1 37) with the nine glutamic acid residues that are carboxylated during posttranslational maturation. A short amphipathic helix (amino acids 38 45) connects the Gla-domain to the first epidermal growth factor (EGF) domain (amino acids 46 91). The second EGF domain (amino acids ) is followed by the activation peptide domain (amino acids ). This region contains the Lys156 Arg157 dipeptide that is released upon maturation, and the cleavage site for thrombin activation (Arg169). The heavy chain of PC contains the activation peptide and the catalytic domain (amino acids ). The gene for human PC (PROC) is located at the 2q13 q14 position and contains nine exons encoding for a 1.7-kb messenger RNA (mrna) and eight introns (10, 11). All the exon/intron boundaries follow the GT-AG rule. Exon 1 is a non-coding exon and a long intron separates it from the initiator ATG codon; this phenomenon is unique among the vitamin K dependent factors. The PROC contains two Alu repeats in intron 5. The major transcriptional start site is located 1515 base pairs upstream from the initiator ATG codon. Two minor transcription start sites were also recognized at 7 and q13 bp relative to the major start site. The regulation of transcription was extensively studied. Cis-elements within PROC are the HNF-1, HNF-3 and Sp1 binding sites. A strong silencer region and two liver specific enhancer regions have also been described. Protein and gene structure of protein S PS is a vitamin K-dependent single-chain 71 kda glycoprotein. It is synthesized primarily in the liver. However, significant amount of PS are also produced by endothelial cells and megakaryocytes (12). Its plasma concentration is mg/l and PS circulates with a half-life of 42 h. Its domain structure is different from other vitamin K dependent proteins. The pre-pro leader sequence (numbered as 41 to 1), the Gla-domain (amino acids 1 36) with 11 glutamic acid residues and a short amphipathic helix (amino acids 37 46) also exists in other vitamin K dependent hemostatic proteins. In addition PS contains unique domains, a thrombin cleavage sensitive loop formed from 24 residues, four EGF domains, and a huge sex hormone domain with the binding site for the complement regulator C4bBP. About two third of PS circulates in complex with C4bBP. In this complex, PS is not a cofactor of APC (13). The gene for human PS (PROS1) is located at position 3q11.2 and contains 15 exons producing a 3.5-kb messenger RNA (mrna) and 14 introns (10). All exon/intron boundaries follow the AG/GT rule. Exon 1 encodes for the first part of the pre-pro leader sequence, exon 2 codes for the second part of the leader sequence and the Gla-domain. Exon 3 encodes for the short hydrophobic region and exon 4 codes for the thrombin sensitive region. Exons 5 8 encode for the four EGF domains. Exons 9 15 code for the very large C4bBP-binding (sex homone binding) domain. Six Alu repeats are located within the PROS1 gene. In addition to the active gene, a pseudogene of PS (PROS2) has also been discovered. This inactive gene shows 97% homology to the active gene, but lacks exon 1 and contains multiple base changes. The presence of PROS2 makes the molecular genetic diagnosis of hereditary PS deficiency a difficult task (see later). Protein C and S deficiencies: epidemiological aspects and clinical symptoms The first patient with PC deficiency was described by Griffin et al. in 1981 (14). PS deficiency was first reported in 1984 by Comp and Esmon (15). Both cases were associated with

61 Bereczky et al.: Protein C and protein S deficiencies S55 recurrent venous thromboembolism. Since then, a large number of deficient patients have been identified and the underlying genetic defects have been clarified in a number of cases (see later). Soon afterwards it was revealed that PC and PS deficiencies are associated with increased thrombotic risk. The degree of the risk and the prevalence of PC and PS deficiencies among patients with thrombosis and in the general population have been examined in several population studies, with conflicting results (16 29). The estimated prevalence of PC deficiency in the general population is surprisingly high (0.4%) (18, 19). The prevalence of PS deficiency seems to be less, but data are uncertain, which is due, at least in part, to difficulties in the laboratory diagnosis. In a Scottish study, it was within a range of 0.03% 0.13% (20), while PS deficiency was shown to be more prevalent in a general Japanese population (1% 2%) (21). Among 163 randomly selected patients with lower extremity venous thrombosis (LEVT), 2% 4% of the patients suffered from PC or PS deficiency (22). Interestingly in the same study, when patients with confirmed cerebral venous sinus thrombosis were examined (ns163), no PC deficiency was found, while the prevalence of PS deficiency did not differ significantly from that of the LEVT group. In an Italian cohort, the prevalence of PC and PS deficiency among patients with first time VTE was found to be 4.7% and 3.7%, respectively (23). PC or PS deficiency was diagnosed in 3% and 2% of patients with proximal deep vein thrombosis, respectively (ns920) (24). According to the results of the Leiden Thrombophilia Study, one of the first case-control studies on this topic, the risk of first VTE in PC deficiency was 3-fold, while in PS deficiency, no significant risk was demonstrated in unselected patients (25). In other studies, a 3 11-fold risk of VTE was demonstrated in PC or PS deficiency. The results depended on the selection of patients, including ethnicity, study design, and the methods for determining PC and PS activity and concentration. The annual incidence of recurrent VTE were 6.0% for PC deficiency and 8.4% for PS deficiency (23, 26). In the large prospective EPCOT study (European Prospective Cohort on Thrombophilia), the risk of first VTE in asymptomatic relatives of patients with confirmed thrombophilia was investigated (ns575). During the 5.7 years of follow-up, 4.5% of individuals developed VTE. The annual incidence of VTE in PC (0.7%) and PS (0.8%) deficiency was higher than in cases of FVL, and thrombosis developed at a mean age of 40 (27). Putting the results of all epidemiological studies together, one can conclude that the risk of developing thrombosis among individuals with genetic defect in PROC or PROS1 varies considerably in the various studies that have been preformed. In addition to methodological variability, this could be due to gene-gene or gene-environment interactions, many of which have not yet been discovered (30, 31). Symptoms of PC or PS deficiencies are deep venous thrombosis and/or pulmonary embolism in early adulthood, which is often recurrent. Thrombosis might also develop at unusual sites, such as the proximal extremities and in mesenterial and cerebral veins. Intracardial thrombus was reported in a 2-year-old child having inherited PC deficiency (32), and intracardial multichamber thrombi were identified in a middle aged patient with combined PC and PS deficiency (33). Pregnancy associated thrombosis has also been reported and PS deficiency was also found in the background of late fetal loss (34). In severe PC and PS deficiency when plasma PC or PS concentrations are extremely low, severe thrombosis develops in newborns, frequently in disseminated form, named purpura fulminans (35). Warfarin-induced skin necrosis is a severe complication of PC or PS deficient patients receiving vitamin K antagonist treatment. In addition to venous thrombosis, patients with PC or PS deficiency can also suffer from thrombotic complications of arterial origin (36). In addition to isolated case reports (37 39), a large cohort of relatives of VTE patients with PC, PS or ATIII deficiency (ns468) had a higher incidence of arterial thrombosis compared to subjects who were not deficient. The risk of arterial thrombosis was especially high in individuals -55 years of age; adjusted hazard ratios for PC and PS deficiencies were 6.9 (95% CI, ) and 4.6 (95% CI, ), respectively (40). In the ARIC cohort which enrolled more than 13,000 patients with coronary events or ischemic stroke, and had a follow-up time of almost 17 years, low PC concentrations were associated with the development of stroke (41). A Japanese study demonstrated that patients with PC deficiency were 10 years younger at the onset of myocardial or cerebral infarction compared with those without deficiency (16). A meta-analysis of studies involving children with arterial ischemic stroke calculated an odds ratios of 8.46 for PC and 3.20 for PS deficiency (42). These findings suggest the importance of screening for inherited thrombophilia, primarily PC and PS deficiencies in young patients with arterial thrombotic events, especially in those without any other obvious risk factor. Molecular genetic background of protein C and S deficiency, genotype-phenotype correlations Protein C deficiency PC deficiency is classified as type I (quantitative) and type II (qualitative) deficiency. In type I deficiency, PC activity and the antigen concentration are decreased equally, suggesting defective synthesis or secretion of the protein, while in type II deficiency, the activity is decreased without a significant decrease in antigen concentrations. The latter type could be due to abnormalities in substrate, calcium-ion or receptor binding. The inheritance of PC deficiency is not as clear as was first thought. It may show an autosomal recessive or dominant inheritance, often with incomplete penetrance. The majority of PC deficient patients are heterozygous for the defect, with typical PC activity values between 30% and 65%. Homozygous or compound heterozygous patients often have undetectable PC concentration and/or activity and exhibit life-threatening thrombosis very early in life. The molecular genetic background of PC deficiency is hetero-

62 S56 Bereczky et al.: Protein C and protein S deficiencies geneous. Most of the mutations cause type I deficiency, type II deficiency is diagnosed in approximately 10% 15% of cases. Summary reports of mutations leading to decreased PC concentrations were first described in 1995 (43, 44). To date, approximately 250 causative mutations have been published, which are collected in different databases ( and (Figure 1). A recently developed mutation database, ProCMD, is an interactive tool that contains phenotypic descriptions with functional and structural data obtained by molecular modeling (47). Most of the mutations causing type I deficiency are single nucleotide substitutions within the coding region of PROC, leading to amino acid changes (approx. 70%). If performed, molecular modeling studies, in almost all cases, suggested misfolding and instability of the mutant proteins as a consequence. In vitro expression studies were performed in approximately one third of the cases only. Point mutations introducing a stop codon were also reported (approx. 5%). A smaller number of the point mutations (approx. 9% of all mutations) were found at the exon/intron boundaries leading to splicing defects. Most of the small deletions (approx. 8%) or insertions (approx. 4%) introduced frameshifts, resulting in a premature stop codon and truncated protein. Gross deletions were identified in only 1% of cases. Although almost all missense mutations result in an absolute block in secretion, a few mutations allow the protein to be secreted. However, the rate of secretion is much lower when compared to the wild type protein (48). In homozygous patients having such mutations, PC concentrations are low, but higher than 1%. Diagnosis of type II deficiency is based on the discrepancy between the results of functional testing and antigen measurements (see later for details). Missense mutations are the most frequently reported types; the resulting amino acid change involving the Gla-domain or the pro-peptide result in defective calcium and phospholipid binding (49 57). Mutations in the serine protease domain result in defective protease activity or decreased substrate binding (58 60). Of interest there are mutations that are enriched in certain populations, suggesting the presence of a founder effect. For example, all Finnish type II protein C deficient cases show a single mutation (p.w380g) (61). The p.r147w mutation within PROC is common in Taiwanese Chinese patients with VTE (62). Five frequent mutations account for almost 50% of all PC deficiencies in patients from Japan (c.1268delg, p.f139v, p.r211w, p.v339m and p.m406i) (63). A common ancestor was identified for probands with the p.r306x mutation in the Dutch population (64). The c.3363insc mutation was introduced by French settlers to North America (65). No mutation in the PROC is detectable in 10% 30% of families with PC deficiency (66). This does not necessarily mean that PROC lacks causative mutations in these cases. The larger gene deletions may remain undetectable when using the classical Sanger (i.e., chain termination) sequencing method for detection of mutations. There are known polymorphisms within the PROC gene which may affect measured PC activity or antigen concentrations (see later for details). This may be regulated by loci other than PROC. As part of the GAIT project, a genome-wide linkage study was performed to localize genes influencing variations in PC plasma concentrations. A region flanked by microsatellite markers D16S3106 and D16S516 on chromosome 16 (16q22 23) with one candidate gene was identified as a major quantitative trait locus influencing variation in PC concentrations. This gene encodes a quinone reductase, NADPH:menadione oxidoreductase 1 (NQO1), involved in vitamin K metabolism. The association of this locus with other vitamin K dependent factor concentrations was also demonstrated, however, the linkage was not as strong as in the case of PC (67). Protein S deficiency Initially three types of PS deficiency were distinguished according to the results of the functional test, and free and total PS antigen concentration measurements. In type I deficiency, a low amount of activity, total and free PS antigen concentrations can be found. In type II deficiency, only the result of the functional test is abnormal, while in type III deficiency, low PS activity is associated with low free PS but normal total PS antigen concentrations. Based on extensive family studies, it was suggested that type I and type III deficiencies are phenotypic variants of the same genetic alteration (68). The molecular background of this finding was that C4BbP and PS could bind to each other with very high affinity. Therefore, in the case of mild PS deficiency, the complexed form of PS is not decreased (69). Later, genetic differences between type I and type III PS deficiency was demonstrated, and no linkage to the PROS1 locus was found in most of the patients having type III deficiency. In families having the PROS1 mutation, the phenotype more often shows type I rather than type III deficiency. These findings led to the conclusion that type I PS deficiency is a monogenic disease caused by PROS1 mutations, while type III PS deficiency is more complex or heterogeneous disorder (70, 71). The majority of protein S-deficient patients are heterozygous for an inherited defect, and homozygous or compound heterozygous deficiency can cause the same symptoms as in the case of PC deficiency. The molecular genetic background of PS deficiency is also heterogeneous (72). Most of the mutations cause type I deficiency, type II deficiency is diagnosed in approximately 5% of cases. The mutations are listed in the HGMD and in the ISTH databases ( hgmd.cf.ac.uk and (Figure 2). Among the 200 different mutations found to date, missense (approx. 53%) mutations are the most frequent. Approximately 20% of the mutations are small deletions or insertions; non-sense and splice-site mutations are present in approximately 14% and 10% of the cases, respectively. Type II PS deficiency is caused by missense mutations affecting the Gla-domain or EGF4 domain (73). Gross deletions are more frequent than in the case of PC deficiency, they comprise approximately 3% of all cases and are associated with quantitative PS deficiency. However, it is very likely that the number of large

63 Bereczky et al.: Protein C and protein S deficiencies S57 Figure 1 Distribution of causative mutations published to date within the PROC gene according to the HGMD ( database. Nucleotide numbering in exon 1 and nearby is given relative to the first nucleotide of the non-coding exon 1. Nucleotide numbering in the coding region is given relative to the first nucleic acid of the initiator ATG codon. Amino acids are numbered according to the mature protein, where the first methionine is numbered as 42. The literature also mentions two gross deletions: one includes the entire gene, the other includes exons 1 9 (45, 46) (not shown in the Figure.).

64 S58 Bereczky et al.: Protein C and protein S deficiencies Figure 2 Distribution of causative mutations published to date within the PROS1 gene according to the HGMD ( database. Nucleotide numbering is given relative to the first nucleic acid of the initiator ATG codon. Amino acids are numbered according to the mature protein, where the first methionine is numbered 41.

65 Bereczky et al.: Protein C and protein S deficiencies S59 deletions in PROS1 is even higher, but often remains undetected. In the PROSIT study, mutations in PROS1 were found only in 70% of probands with PS deficiency (17). However, using DNA sequencing methods, large deletions or gene segment duplications may remain undetected (74, 75). Moreover, there are mutations affecting the transcription regulatory sequences at the 5 of the gene (76, 77). Most recently, the first case of PS deficiency due to chromosome translocation has been reported. The diagnosis was established by painting fluorescence in situ hybridization (78). Although a founder effect is not confirmed, it is to be noted that PS Tokushima (p.k155e) shows a high prevalence in the Japanese population. Polymorphisms in the protein C and S gene It has been reported that individuals with the homozygous C/G/T haplotype at the nucleotide positions (rs ), (rs ), and (rs ) in the promoter region of the PROC had lower plasma PC concentrations compared to individuals with the T/A/A homozygous haplotype (79). The -1654/-1641 CC/GG genotype was associated with a slightly increased risk of thrombosis (OR, 1.39, 95% CI, ) (80). In a large populationbased case-control study, individuals having this genotype had the lowest plasma PC concentration, and the highest risk for venous thrombosis (OR, 1.27, 95% CI, ) compared with individuals having the TT/AA genotype (81). In contrast to these findings, variation at the PROC structural locus did not influence plasma PC concentrations in the GAIT project, but the chromosome region 16q22-23 was found to be a major determinant (67). Polymorphisms in genes involved in the vitamin K dependent g-carboxylation of PC and PS may also be responsible for the inter-individual variation in the plasma concentrations of these proteins in the general population (82). PS Heerlen (p.s460p) results in the loss of N-glycosylation at Asn458. The concentration of free PS in the plasma of carriers was slightly lower than that of non-carriers and was considered to be a type III PS deficiency. However, the risk of thrombosis conferred by this mutation is a matter of debate. PS Heerlen displayed reduced anticoagulant activity as cofactor to APC in plasma based assays, as well as when using a FVIIIa degradation system. In a purified system using recombinant proteins, PS Heerlen was a good cofactor of APC in the degradation of normal FVa, but became a poor cofactor in the degradation of FVa carrying the Leiden mutation (83). This suggested a synergistic contribution between FV Leiden and PS Heerlen that increases the risk of thrombosis. However, this hypothesis was not confirmed by another study. The cause of the decrease in free PS concentrations associated with PS Heerlen has not been clarified, but most likely is a consequence of increased clearance (84). A transition of adenine to guanine transition at nt 2148 (p.p626p, silent) and an A to C substitution at nt 2698 have been suggested to decrease PS concentration in healthy individuals (85). However, this finding was not confirmed in another study (86). No decrease in the secretion of p.p626p variant was demonstrated in an in vitro expression system; this variant was not a risk factor for VTE and did not modify the risk of patients with causative mutations (87). Laboratory tests of protein C deficiency Two different types of assays are available for the diagnosis and classification of PC deficiency, functional tests and antigen assays. For screening, a functional test should be performed, and if the results are abnormal, the antigen assay can distinguish between type I and type II deficiencies, with concentrations of antigen being normal in the latter. There are two different methods for determination of PC activity (Figure 3A). In both assays, PC present in patient plasma is activated by the venom of Agkistrodon contortrix, now commercially available under the trade name Protac. The advantage of Protac is its insensitivity to plasma protease inhibitors. Also, it can be added directly to plasma. Protac activated PC can be measured either using a chromogenic assay or a clotting assay. In chromogenic assays, paranitroaniline (pna) is cleaved-off from a small synthetic peptide by APC. Peptide bound pna does not absorb light at 405 nm, while the liberated chromogenic compound has an intense color at this wavelength. The spectrophotometric measurement can be either end-point or kinetic, with the latter being preferred. The higher the APC activity, the more intense the increase in absorbance during the test. The rationale of the clotting time based assays is the fact that if APC degrades its natural substrates FVa and FVIIIa, it leads to clotting time prolongation. Determination of clotting time can be based on the prothrombin time, activated thromboplastin time (APTT) or Russell viper venom time (RVV), and there is a linear relationship between PC activity and clotting time. There are numerous advantages and disadvantages of both functional assays (88) (Table 1). Clotting tests are influenced by several pre-analytical or analytical variables. Despite predilution of the sample with PC deficient plasma, in the presence of lupus anticoagulant, heparin or direct thrombin (or factor Xa) inhibitor which prolongs the clotting time, falsely increased PC activity can be measured (89). Interference by heparin (up to 1 2 U/mL) is eliminated by adding a heparinneutralizing substance, e.g., hexadimethrine bromide (polybrene) to the reagent. The interference caused by lupus anticoagulant is more pronounced in tests using APTT as activator (90). In the case of increased Factor VIII, the opposite effect might be seen; shortening the clotting time (APTT) will lead to falsely decreased PC activity (91). The most important problem with the clotting method is the influence of the FV Leiden mutation. In patients having this mutation, falsely low PC activity could be detected despite pre-dilution of patient plasma with PC deficient plasma containing wild type FV. If PC antigen concentrations are normal, antigen measurements are not influenced by the mutation; such patients are easily misdiagnosed as having type II PC deficiency (92, 93).

66 S60 Bereczky et al.: Protein C and protein S deficiencies Figure 3 Schematic presentations of protein C and protein S functional assays. PC, protein C; PS, protein S; APC, activated protein C; R-pNA, chromogenic substrate containing oligopeptide (R) and p-nitroaniline (pna); DA, difference in absorbance. Chromogenic assays are not sensitive to high FVIII, lupus anticoagulant or FV Leiden. They show lower inter-laboratory and intra-laboratory variation (94). There are two major limitations in the chromogenic assays; the chromogenic peptide substrates do not have exclusive specificity for APC and may overestimate PC in the presence of other proteolytic enzymes, such as plasmin, kallicrein, and thrombin which also cleave the chromogenic substrate(s). The second problem is that chromogenic assays are insensitive to a certain type of qualitative PC deficiency. When the mutation affects Table 1 The most important difficulties in the diagnosis of protein C and protein S deficiency. Analytical/methodical problems Protein C Protein S Functional clotting assays Lupus anticoagulant Overestimation High heparin concentration Overestimation Direct thrombin (or FXa) inhibitor Overestimation High FVIII level (usually )250%) Underestimation FV Leiden mutation Underestimation Functional amidolytic assay Presence of enzymes cleaving the Overestimation NA chromogenic substrate Mutations result in altered Normal result despite genetic defect NA g-carboxylation, or phospholipid binding Problems with molecular genetic diagnosis Large gene segment alterations are not diagnosed by the DNA sequencing method NA Presence of the pseudogene (PROSP) causes difficulties Physiological conditions which influence PC-PS levels Pregnancy Significant elevation in the first 22 weeks of pregnancy Decreased Oral anticoncipients, hormonal Decreased replacement therapy Infants Levels are significantly lower than adult values at birth and infancy Age Increases with age Gender Lower in women Acquired deficiency Vitamin K antagonist therapy Decreased Hepatic disease Decreased Consumption (DIC, VTE) Decreased Presence of autoantibodies (SLE, Decreased varicella, malignancies, sepsis, HIV)

67 Bereczky et al.: Protein C and protein S deficiencies S61 the Gla-domain or the propeptide, the functional clotting test gives a low value for activity, while the amidolytic (chromogenic) assay shows normal result. Mutations in the propeptide affecting the g-carboxylation process, substitution of a Gla residue, or mutations influencing phospholipid binding decrease PC activity as measured using the clotting test, while amidolytic activity remains unaltered (49 54). In general, PC deficiency caused by mutations in the serine protease domain can be diagnosed using both clotting and amidolytic assays. However, some mutations in this region (p.r229q and p.s252n) result in abnormalities that can only be demonstrated with the clotting assay. It is estimated that approximately 5% of patients with type II PC deficiency have normal amidolytic activity, and the deficiency is detectable only by clotting assays. Initially PC antigen was measured using electroimmunoassay. However, this assay is now considered obsolete and no longer routinely used. Most laboratories perform commercially available ELISA testing using pairs of monoclonal or polyclonal antibodies against PC. Laboratory tests of protein S deficiency Three types of assays are available for the determination of plasma PS: the functional PS activity assay, free and total PS antigen determinations. The test for PS activity measures the effect of PS as a cofactor on the degradation of FVa and FVIIIa by APC. Such clotting tests are the function of free PS concentrations; they are not influenced by PS in complex with C4bBP (Figure 3B). Commercially available tests use thromboplastin, APTT, RVV or FXa for initiation of coagulation. APC is added to patient or control plasma that has been pre-diluted with PS deficient plasma. Following this, the clotting time test is performed. The effect of APC on clotting time, i.e., the extent that the clotting time is prolonged, depends on the content of the cofactor PS in the plasma to be investigated. PS activity assays have a number of limitations (Table 1). First, in patients having FV Leiden mutation, significantly lower PS activity is measured, the results often overlap with the range for true PS deficiency. Such situations may lead to an incorrect diagnosis of type II PS deficiency (95). In the majority of assays, purified FVa is added to the test. This step decreases somewhat the interference from FV Leiden, but does not eliminate the problem completely. Different kits give highly variable results; they have different cut-off values and most of them are very sensitive to reagent handling. Pre-analytical variables are similar to those described for PC activity measurements; high FVIII may lead to underestimation, while the presence of lupus anticoagulant may lead to overestimation of PS activity. Plasma samples are sensitive to repeated freezing and thawing; in vitro activation of FVII may shorten the clotting time in assays using thromboplastin as activator, resulting in underestimation of PS activity. Commercially available PS activity assays give the correct diagnosis of PS deficiency in 97% of patients having PROS1 mutations (96). However, the specificity of the functional tests is low due to the above-mentioned interfering factors that might lead to inappropriate interpretations of test results, resulting in a false-positive diagnosis. For this reason, some authors do not recommend the use of PS activity assays for diagnosis of PS deficiency, and instead favor free PS antigen (see below) determinations (97, 98). However, by omitting the functional assay, type II deficient patients would remain undiagnosed. Therefore, it has been suggested that both activity and free antigen assays be perfomed from the same sample (99). For the measurement of total PS concentrations, ELISA is the most frequently used method. Measurement of free PS antigen was originally performed from the supernatant following precipitation of C4bBP-bound PS by polyethylene glycol. This method was time consuming and poorly reproducible. Later, monoclonal antibodies against the C4bBP binding domain of PS were produced and measurement of free PS antigen became easier and faster using ELISA assays. Further development led to the introduction of latex enhanced immunoassays (LIA) which were easily adapted to automated coagulometers (100). In the latest ECAT exercise, 77% of laboratories used a LIA method for free PS antigen measurements. In addition to monoclonal antibodies specific for free PS, a ligand binding assay has also been developed. In this assay, C4bBP is used to capture free PS from the plasma (101). Assays for the free form of PS are preferred over total PS determinations since this has higher positive predictive value for PS deficiency (68, 69). In the latest exercise of the European Concerted Action on Thrombosis (ECAT) thrombophilia testing program, only 89 laboratories reported total PS antigen results and 225 laboratories performed free PS antigen measurements. Conditions affecting protein C and protein S levels; acquired deficiencies Adult reference intervals for PC and PS are wide, and there may be overlap between values seen in healthy individuals and deficient patients (102). In infants, both PC and PS concentrations are lower than adult values. In a healthy full term infant, PC activity is 35% (17% 53%), PS activity is 36% (24% 48%); these reach the lower limit of the adult reference interval by 1 year of age. The PC concentration may remain below the adult reference interval until adolescence ( ). It should also be noted that PS concentrations are influenced by age and gender; lower results are obtained in women compared with men, and PS values increase with age. PS concentrations may decrease markedly during pregnancy, to a mean level of 46%, and to a lesser extent in individuals using oral contraceptives or on hormone replacement therapy (107, 108). PS measurements in pregnant women can only be used as a test for exclusion, decreased PS concentrations cannot be considered as being deficient. During the first 22 weeks of pregnancy, PC concentrations show a significant increase. It has been postulated that this increase may play a role in maintaining early pregnancy by regulating both coagulation and inflammation (107).

68 S62 Bereczky et al.: Protein C and protein S deficiencies Treatment of patients with vitamin K antagonist (VKA) therapy influences plasma PC and PS activity and antigen concentrations. Activities of PC and PS are markedly decreased and, depending on the assay, antigen may also be lower. Using typical therapeutic doses of VKA, PC antigen and activity decreases to approximately 50% and 25%, respectively (109). Patients should not receive VKA therapy for at least 10 days prior to testing, they should be switched to low molecular weight heparin therapy until collecting the sample for PC and PS measurements. As the half-life of PC is much more shorter than that of PS, it decreases faster following the initiation of VKA therapy and recovers more rapidly following discontinuation. Abnormal g-carboxylation due to vitamin K deficiency also results in decreased PC and PS activity and antigen concentrations. PC and PS deficiency can develop with DIC, severe infection, sepsis and acute excessive thrombosis due to consumption (110, 111). Decreased synthesis of PC and PS can be a consequence of liver disease or immaturity of the liver in preterm infants. Autoimmune syndromes can also be associated with acquired PC and PS deficiency due to the presence of autoantibodies. Postvaricella purpura fulminans is a rare complication in children caused by acquired PC or PS deficiency (112). PS deficiency has also been described in patients with AIDS (113, 114). Therapy with L-asparaginase may lead to decreased PC concentrations by decreasing its synthesis in the liver. Patients having nephrotic syndrome may also exhibit low PS concentrations. Molecular genetic diagnosis of protein C and protein S deficiencies Since both PC and PS deficiencies may be acquired. Prior to suggesting a genetic defect, all the possible acquired conditions must be excluded. Equivocal cases require confirming the presence of a true inherited deficiency using mutation analysis. As multiple sites of mutation have already been described in PROC and PROS1 genes, and since no so-called hot spot could be identified within these genes, DNA sequencing is the most reliable method for establishing a genetic diagnosis. The molecular genetic diagnosis of PS deficiency represents a particularly difficult situation. The presence of the PS pseudogene makes the genetic diagnosis of PS deficiency rather complicated; careful design of primers are required to eliminate the amplification of pseudogene fragments. In a high number of individuals with PS deficiency, no mutation was found when using DNA sequencing. This discrepancy is due to larger gene alterations that are not diagnosed by this method (66, 75, 115). A recently developed and commercially available method for demonstrating large gene segment deletions or duplications is the multiplex ligationdependent probe amplification (MLPA) method. Re-analysis of DNA samples from mutation negative PS deficient patients using this method has revealed large deletions or duplications in a number of cases (74). The presence of such larger gene alterations should be confirmed by other methods, such as quantitative PCR or long PCR. Concluding remarks The diagnosis of PC and PS deficiency is not an easy task; the functional tests are influenced by several pre-analytical and analytical factors, and the molecular genetic diagnosis is also challenging. In the case of PC, the use of the chromogenic or the clotting functional test as screening tests is a matter of debate. A clinical guideline most recently issued by UK-based medical experts recommends the chromogenic PC assay as being the preferred test (116). In our experience, only the use of both chromogenic and clotting tests could cover the full range of PC deficiencies and reduce the problems arising from interfering conditions. In the diagnosis of PS deficiency, if a PS activity assay is used for initial screening, low results should be further investigated using a immunoreactive assay for free PS. Acquired deficiencies should be considered and looked for when establishing the diagnosis. In all cases, repeat testing is crucial for establishing the diagnosis. If the results of functional and antigenic assays do not confirm the diagnosis unequivocally, genetic testing is indicated. Acknowledgements This work was supported by a grant from the Hungarian National Research Fund (OTKA K78386). Conflict of interest statement Author s conflict of interest disclosure: The authors stated that there are no conflicts of interest regarding the publication of this article. Research funding: None declared. Employment or leadership: None declared. Honorarium: None declared. References 1. Reitsma PH, Rosendaal FR. Past and future of genetic research in thrombosis. J Thromb Haemost 2007;5Suppl 1: Dahlback B. Factor V and protein S as cofactors to activated protein C. Haematologica 1997;82: Dahlback B. Advances in understanding pathogenic mechanisms of thrombophilic disorders. Blood 2008;112: Esmon NL, Smirnov MD, Safa O, Esmon CT. Lupus anticoagulants, thrombosis and the protein C system. Haematologica 1999;84: Mosnier LO, Zlokovic BV, Griffin JH. The cytoprotective protein C pathway. Blood 2007;109: Heeb MJ, Mesters RM, Tans G, Rosing J, Griffin JH. Binding of protein S to factor Va associated with inhibition of prothrombinase that is independent of activated protein C. J Biol Chem 1993;268:

69 Bereczky et al.: Protein C and protein S deficiencies S63 7. van Wijnen M, van t Veer C, Meijers JC, Bertina RM, Bouma BN. A plasma coagulation assay for an activated protein C- independent anticoagulant activity of protein S. Thromb Haemost 1998;80: Hackeng TM, Sere KM, Tans G, Rosing J. Protein S stimulates inhibition of the tissue factor pathway by tissue factor pathway inhibitor. Proc Natl Acad Sci USA 2006;103: Tuddenham EG, Cooper DN. Protein C and protein C inhibitor. In: Tuddenham EG, Cooper DN, editors. The molecular genetics of haemostasis and its inherited disorders. Oxford: Oxford University Press, 1994: Esmon CT. Protein C, protein S, and thrombomodulin. In: Colman RW, Marder VJ, Clowes AW, George JN, Goldhaber SZ, editors. Hemostasis and thrombosis, basic principles and clinical practice. Philadelphia: Lippincott and Williams and Wilkins, 2006: Patracchini P, Aiello V, Palazzi P, Calzolari E, Bernardi F. Sublocalization of the human protein C gene on chromosome 2q13-q14. Hum Genet 1989;81: Tuddenham EG, Cooper DN. Protein S, C4b-binding protein, and protein Z. In: Tuddenham EG, Cooper DN, editors. The molecular genetics of haemostasis and its inherited disorders. Oxford: Oxford University Press, 1994: Dahlback B, Stenflo J. High molecular weight complex in human plasma between vitamin K-dependent protein S and complement component C4b-binding protein. Proc Natl Acad Sci USA 1981;78: Griffin JH, Evatt B, Zimmerman TS, Kleiss AJ, Wideman C. Deficiency of protein C in congenital thrombotic disease. J Clin Invest 1981;68: Comp PC, Esmon CT. Recurrent venous thromboembolism in patients with a partial deficiency of protein S. N Engl J Med 1984;311: Sakata T, Kario K, Katayama Y, Matsuyama T, Kato H, Miyata T. Studies on congenital protein C deficiency in Japanese: prevalence, genetic analysis, and relevance to the onset of arterial occlusive diseases. Semin Thromb Hemost 2000;26: Biguzzi E, Razzari C, Lane DA, Castaman G, Cappellari A, Bucciarelli P, et al. Molecular diversity and thrombotic risk in protein S deficiency: the PROSIT study. Hum Mutat 2005;25: Miletich J, Sherman L, Broze G, Jr. Absence of thrombosis in subjects with heterozygous protein C deficiency. N Engl J Med 1987;317: Tait RC, Walker ID, Reitsma PH, Islam SI, McCall F, Poort SR, et al. Prevalence of protein C deficiency in the healthy population. Thromb Haemost 1995;73: Dykes AC, Walker ID, McMahon AD, Islam SI, Tait RC. A study of protein S antigen levels in 3788 healthy volunteers: influence of age, sex and hormone use, and estimate for prevalence of deficiency state. Br J Haematol 2001;113: Miyata T, Kimura R, Kokubo Y, Sakata T. Genetic risk factors for deep vein thrombosis among Japanese: importance of protein S K196E mutation. Int J Hematol 2006;83: Wysokinska EM, Wysokinski WE, Brown RD, Karnicki K, Gosk-Beirska I, Grill D, et al. Thrombophilia differences in cerebral venous sinus and lower extremity deep venous thrombosis. Neurology 2008;70: De Stefano V, Simioni P, Rossi E, Tormene D, Za T, Pagnan A, et al. The risk of recurrent venous thromboembolism in patients with inherited deficiency of natural anticoagulants antithrombin, protein C and protein S. Haematologica 2006;91: Rossi E, Za T, Ciminello A, Leone G, De Stefano V. The risk of symptomatic pulmonary embolism due to proximal deep venous thrombosis differs in patients with different types of inherited thrombophilia. Thromb Haemost 2008;99: Koster T, Rosendaal FR, Briet E, van der Meer FJ, Colly LP, Trienekens PH, et al. Protein C deficiency in a controlled series of unselected outpatients: an infrequent but clear risk factor for venous thrombosis (Leiden Thrombophilia Study). Blood 1995; 85: Brouwer JL, Lijfering WM, Ten Kate MK, Kluin-Nelemans HC, Veeger NJ, van der Meer J. High long-term absolute risk of recurrent venous thromboembolism in patients with hereditary deficiencies of protein S, protein C or antithrombin. Thromb Haemost 2009;101: Vossen CY, Conard J, Fontcuberta J, Makris M, van der Meer FJ, Pabinger I, et al. Risk of a first venous thrombotic event in carriers of a familial thrombophilic defect. The European Prospective Cohort on Thrombophilia (EPCOT). J Thromb Haemost 2005;3: Shen MC, Lin JS, Tsay W. Protein C and protein S deficiencies are the most important risk factors associated with thrombosis in Chinese venous thrombophilic patients in Taiwan. Thromb Res 2000;99: Altinisik J, Ates O, Ulutin T, Cengiz M, Buyru N. Factor V Leiden, prothrombin G20210A, and protein C mutation frequency in Turkish venous thrombosis patients. Clin Appl Thromb Hemost 2008;14: Tjeldhorn L, Sandset PM, Haugbro K, Skretting G. Hereditary protein C deficiency caused by the Ala267Thr mutation in the protein C gene is associated with symptomatic and asymptomatic venous thrombosis. Thromb Res 2010;125: Scott BT, Bovill EG, Callas PW, Hasstedt SJ, Leppert MF, Valliere JE, et al. Genetic screening of candidate genes for a prothrombotic interaction with type I protein C deficiency in a large kindred. Thromb Haemost 2001;85: Matitiau A, Tabachnik E, Sthoeger D, Birk E. Thrombus in the left ventricle of a child with systemic emboli: an unusual presentation of hereditary protein C deficiency. Pediatrics 2001; 107: Kim MJ, Hur SH, Lee YS, Hyun DW, Han SW, Kim KS, et al. Intracardiac multichamber thrombi in a patient with combined protein C and protein S deficiencies. Int J Cardiol 2005;100: Pabinger I, Grafenhofer H. Thrombosis during pregnancy: risk factors, diagnosis and treatment. Pathophysiol Haemost Thromb 2002;32: Goldenberg NA, Manco-Johnson MJ. Protein C deficiency. Haemophilia 2008;14: Kohler J, Kasper J, Witt I, von Reutern GM. Ischemic stroke due to protein C deficiency. Stroke 1990;21: Acar G, Dogan A, Altinbas A, Turker Y. Recurrent acute stent thrombosis associated with protein C and S deficiencies. Int J Cardiovasc Imaging 2006;22: Cakir O, Ayyildiz O, Oruc A, Eren N. A young adult with coronary artery and jugular vein thrombosis: a case report of combined protein S and protein C deficiency. Heart Vessels 2002;17: Peterman MA, Roberts WC. Syndrome of protein C deficiency and anterior wall acute myocardial infarction at a young age from a single coronary occlusion with otherwise normal coronary arteries. Am J Cardiol 2003;92: Mahmoodi BK, Brouwer JL, Veeger NJ, van der Meer J. Hereditary deficiency of protein C or protein S confers increased risk

70 S64 Bereczky et al.: Protein C and protein S deficiencies of arterial thromboembolic events at a young age: results from a large family cohort study. Circulation 2008;118: Folsom AR, Ohira T, Yamagishi K, Cushman M. Low protein C and incidence of ischemic stroke and coronary heart disease: the Atherosclerosis Risk in Communities (ARIC) Study. J Thromb Haemost 2009;7: Kenet G, Lutkhoff LK, Albisetti M, Bernard T, Bonduel M, Brandao L, et al. Impact of thrombophilia on risk of arterial ischemic stroke or cerebral sinovenous thrombosis in neonates and children: a systematic review and meta-analysis of observational studies. Circulation 2010;121: Reitsma PH. Protein C deficiency: summary of the 1995 database update. Nucleic Acids Res 1996;24: Aiach M, Gandrille S, Emmerich J. A review of mutations causing deficiencies of antithrombin, protein C and protein S. Thromb Haemost 1995;74: Matsuda M, Sugo T, Sakata Y, Murayama H, Mimuro J, Tanabe S, et al. A thrombotic state due to an abnormal protein C. N Engl J Med 1988;319: Millar DS, Johansen B, Berntorp E, Minford A, Bolton-Maggs P, Wensley R, et al. Molecular genetic analysis of severe protein C deficiency. Hum Genet 2000;106: D Ursi P, Marino F, Caprera A, Milanesi L, Faioni EM, Rovida E. ProCMD: a database and 3D web resource for protein C mutants. BMC Bioinformatics 2007;8 Suppl 1:S Sugahara Y, Miura O, Hirosawa S, Aoki N. Compound heterozygous protein C deficiency caused by two mutations, Arg- 178 to Gln and Cys-331 to Arg, leading to impaired secretion of mutant protein C. Thromb Haemost 1994;72: Bovill EG, Tomczak JA, Grant B, Bhushan F, Pillemer E, Rainville IR, et al. Protein CVermont: symptomatic type II protein C deficiency associated with two GLA domain mutations. Blood 1992;79: Gandrille S, Alhenc-Gelas M, Gaussem P, Aillaud MF, Dupuy E, Juhan-Vague I, et al. Five novel mutations located in exons III and IX of the protein C gene in patients presenting with defective protein C anticoagulant activity. Blood 1993;82: Ido M, Ohiwa M, Hayashi T, Nishioka J, Hatada T, Watanabe Y, et al. A compound heterozygous protein C deficiency with a single nucleotide G deletion encoding Gly-381 and an amino acid substitution of Lys for Gla-26. Thromb Haemost 1993; 70: Gaussem P, Gandrille S, Duchemin J, Emmerich J, Alhenc- Gelas M, Aillaud MF, et al. Influence of six mutations of the protein C gene on the Gla domain conformation and calcium affinity. Thromb Haemost 1994;71: Lu D, Bovill EG, Long GL. Molecular mechanism for familial protein C deficiency and thrombosis in protein CVermont (Glu20--)Ala and Val34--)Met). J Biol Chem 1994;269: Faioni EM, Hermida J, Rovida E, Razzari C, Asti D, Zeinali S, et al. Type II protein C deficiency: identification and molecular modelling of two natural mutants with low anticoagulant and normal amidolytic activity. Br J Haematol 2000;108: Preston RJ, Morse C, Murden SL, Brady SK, O Donnell JS, Mumford AD. The protein C omega-loop substitution Asn2Ile is associated with reduced protein C anticoagulant activity. Br J Haematol 2009;144: Simioni P, Kalafatis M, Tormene D, Luni S, Zerbinati P, Barzon L, et al. Abnormal propeptide processing resulting in the presence of two abnormal species of protein C in plasma: characterization of the dysfunctional protein C Padua3 (protein C(R-1L/propeptide)). Thromb Haemost 2001;86: Nakabayashi T, Mizukami K, Naitoh S, Takeda M, Shikamoto Y, Nakagawa T, et al. Protein C Sapporo (protein C Glu 25 --) Lys): a heterozygous missense mutation in the Gla domain provides new insight into the interaction between protein C and endothelial protein C receptor. Thromb Haemost 2005;94: Marchetti G, Patracchini P, Gemmati D, Castaman G, Rodeghiero F, Wacey A, et al. Symptomatic type II protein C deficiency caused by a missense mutation (Gly 381--)Ser) in the substrate-binding pocket. Br J Haematol 1993;84: Doig RG, Begley CG, McGrath KM. Hereditary protein C deficiency associated with mutations in exon IX of the protein C gene. Thromb Haemost 1994;72: Zheng YZ, Sakata T, Matsusue T, Umeyama H, Kato H, Miyata T. Six missense mutations associated with type I and type II protein C deficiency and implications obtained from molecular modelling. Blood Coagul Fibrinolysis 1994;5: Levo A, Kuismanen K, Holopainen P, Vahtera E, Rasi V, Krusius T, et al. Single founder mutation (W380G) in type II protein C deficiency in Finland. Thromb Haemost 2000;84: Tsay W, Shen MC. R147W mutation of PROC gene is common in venous thrombotic patients in Taiwanese Chinese. Am J Hematol 2004;76: Miyata T, Sakata T, Yasumuro Y, Okamura T, Katsumi A, Saito H, et al. Genetic analysis of protein C deficiency in nineteen Japanese families: five recurrent defects can explain half of the deficiencies. Thromb Res 1998;92: Zeegers MP, van Poppel F, Vlietinck R, Spruijt L, Ostrer H. Founder mutations among the Dutch. Eur J Hum Genet 2004;12: Couture P, Bovill EG, Demers C, Simard J, Delage R, Scott BT, et al. Evidence of a founder effect for the protein C gene 3363 inserted C mutation in thrombophilic pedigrees of French origin. Thromb Haemost 2001;86: Koeleman BP, Reitsma PH, Bertina RM. Familial thrombophilia: a complex genetic disorder. Semin Hematol 1997;34: Buil A, Soria JM, Souto JC, Almasy L, Lathrop M, Blangero J, et al. Protein C levels are regulated by a quantitative trait locus on chromosome 16: results from the Genetic Analysis of Idiopathic Thrombophilia (GAIT) Project. Arterioscler Thromb Vasc Biol 2004;24: Zoller B, Garcia de Frutos P, Dahlback B. Evaluation of the relationship between protein S and C4b-binding protein isoforms in hereditary protein S deficiency demonstrating type I and type III deficiencies to be phenotypic variants of the same genetic disease. Blood 1995;85: Simmonds RE, Ireland H, Lane DA, Zoller B, Garcia de Frutos P, Dahlback B. Clarification of the risk for venous thrombosis associated with hereditary protein S deficiency by investigation of a large kindred with a characterized gene defect. Ann Intern Med 1998;128: Espinosa-Parrilla Y, Morell M, Souto JC, Tirado I, Fontcuberta J, Estivill X, et al. Protein S gene analysis reveals the presence of a cosegregating mutation in most pedigrees with type I but not type III PS deficiency. Hum Mutat 1999;14: Ten Kate MK, Platteel M, Mulder R, Terpstra P, Nicolaes GA, Reitsma PH, et al. PROS1 analysis in 87 pedigrees with hereditary protein S deficiency demonstrates striking genotype-phenotype associations. Hum Mutat 2008;29: Garcia de Frutos P, Fuentes-Prior P, Hurtado B, Sala N. Molecular basis of protein S deficiency. Thromb Haemost 2007;98:

71 Bereczky et al.: Protein C and protein S deficiencies S Baroni M, Mazzola G, Kaabache T, Borgel D, Gandrille S, Vigano D Angelo S, et al. Molecular bases of type II protein S deficiency: the I203-D204 deletion in the EGF4 domain alters GLA domain function. J Thromb Haemost 2006;4: Pintao MC, Garcia AA, Borgel D, Alhenc-Gelas M, Spek CA, de Visser MC, et al. Gross deletions/duplications in PROS1 are relatively common in point mutation-negative hereditary protein S deficiency. Hum Genet 2009;126: Yin T, Takeshita S, Sato Y, Sakata T, Shin Y, Honda S, et al. A large deletion of the PROS1 gene in a deep vein thrombosis patient with protein S deficiency. Thromb Haemost 2007;98: Tatewaki H, Tsuda H, Kanaji T, Yokoyama K, Hamasaki N. Characterization of the human protein S gene promoter: a possible role of transcription factors Sp1 and HNF3 in liver. Thromb Haemost 2003;90: de Wolf CJ, Cupers RM, Bertina RM, Vos HL. The constitutive expression of anticoagulant protein S is regulated through multiple binding sites for Sp1 and Sp3 transcription factors in the protein S gene promoter. J Biol Chem 2006;281: Hurtado B, Nadal M, Margarit E, Sanchez A, Abasolo N, Garcia N, et al. First case of protein S deficiency due to a translocation t(3;21)(q11.2;q22). Thromb Haemost 2009;101: Spek CA, Koster T, Rosendaal FR, Bertina RM, Reitsma PH. Genotypic variation in the promoter region of the protein C gene is associated with plasma protein C levels and thrombotic risk. Arterioscler Thromb Vasc Biol 1995;15: Aiach M, Nicaud V, Alhenc-Gelas M, Gandrille S, Arnaud E, Amiral J, et al. Complex association of protein C gene promoter polymorphism with circulating protein C levels and thrombotic risk. Arterioscler Thromb Vasc Biol 1999;19: Pomp ER, Doggen CJ, Vos HL, Reitsma PH, Rosendaal FR. Polymorphisms in the protein C gene as risk factor for venous thrombosis. Thromb Haemost 2009;101: Kimura R, Kokubo Y, Miyashita K, Otsubo R, Nagatsuka K, Otsuki T, et al. Polymorphisms in vitamin K-dependent gamma-carboxylation-related genes influence interindividual variability in plasma protein C and protein S activities in the general population. Int J Hematol 2006;84: Giri TK, Yamazaki T, Sala N, Dahlback B, de Frutos PG. Deficient APC-cofactor activity of protein S Heerlen in degradation of factor Va Leiden: a possible mechanism of synergism between thrombophilic risk factors. Blood 2000;96: Koenen RR, Gomes L, Tans G, Rosing J, Hackeng TM. The Ser460Pro mutation in recombinant protein S Heerlen does not affect its APC-cofactor and APC-independent anticoagulant activities. Thromb Haemost 2004;91: Leroy-Matheron C, Duchemin J, Levent M, Gouault-Heilmann M. Genetic modulation of plasma protein S levels by two frequent dimorphisms in the PROS1 gene. Thromb Haemost 1999; 82: Heinikari T, Huoponen O, Partanen J, Rasi V, Krusius T. Protein S gene polymorphisms Pro626 and nt2698--no correlation to free protein S levels or protein S activities. Thromb Haemost 2005;94: Castaman G, Biguzzi E, Razzari C, Tosetto A, Fontana G, Asti D, et al. Association of protein S p.pro667pro dimorphism with plasma protein S levels in normal individuals and patients with inherited protein S deficiency. Thromb Res 2007;120: Khor B, Van Cott EM. Laboratory tests for protein C deficiency. Am J Hematol 2010;85: Simioni P, Lazzaro A, Zanardi S, Girolami A. Spurious protein C deficiency due to antiphospholipid antibodies. Am J Hematol 1991;36: Cooper PC, Cooper SM, Goodfellow KJ, Hickey KP, Kitchen S, Makris M. Evaluation of a new venom-based clotting assay of protein C. Int J Lab Hematol 2008;30: de Moerloose P, Reber G, Bouvier CA. Spuriously low levels of protein C with a Protac activation clotting assay. Thromb Haemost 1988;59: Ireland H, Bayston T, Thompson E, Adami A, Goncalves C, Lane DA, et al. Apparent heterozygous type II protein C deficiency caused by the factor V 506 Arg to Gln mutation. Thromb Haemost 1995;73: Jennings I, Kitchen S, Cooper PC, Rimmer JE, Woods TA, Preston FE. Further evidence that activated protein C resistance affects protein C coagulant activity assays. Thromb Haemost 2000;83: Meijer P, Kluft C, Haverkate F, De Maat MP. The long-term within- and between-laboratory variability for assay of antithrombin, and proteins C and S: results derived from the external quality assessment program for thrombophilia screening of the ECAT Foundation. J Thromb Haemost 2003;1: Deitcher SR, Kottke-Marchant K. Pseudo-protein S deficiency due to activated protein C resistance. Thromb Res 2003;112: Makris M, Leach M, Beauchamp NJ, Daly ME, Cooper PC, Hampton KK, et al. Genetic analysis, phenotypic diagnosis, and risk of venous thrombosis in families with inherited deficiencies of protein S. Blood 2000;95: De Stefano V, Finazzi G, Mannucci PM. Inherited thrombophilia: pathogenesis, clinical syndromes, and management. Blood 1996;87: Lane DA, Mannucci PM, Bauer KA, Bertina RM, Bochkov NP, Boulyjenkov V, et al. Inherited thrombophilia: Part 1. Thromb Haemost 1996;76: Haemostasis and Thrombosis Task Force, British Committee for Standards in Haematology. Investigation and management of heritable thrombophilia. Br J Haematol 2001;114: Deffert C, Esteve F, Grimaux M, Gouault-Heilmann M. A direct, automated, immuno-turbidimetric assay of free protein S antigen in plasma. Blood Coagul Fibrinolysis 2001;12: Giri TK, Hillarp A, Hardig Y, Zoller B, Dahlback B. A new direct, fast and quantitative enzyme-linked ligand sorbent assay for measurement of free protein S antigen. Thromb Haemost 1998;79: Finazzi G, Barbui T. Different incidence of venous thrombosis in patients with inherited deficiencies of antithrombin III, protein C and protein S. Thromb Haemost 1994;71: Andrew M, Paes B, Milner R, Johnston M, Mitchell L, Tollefsen DM, et al. Development of the human coagulation system in the full-term infant. Blood 1987;70: Andrew M, Vegh P, Johnston M, Bowker J, Ofosu F, Mitchell L. Maturation of the hemostatic system during childhood. Blood 1992;80: Petaja J, Manco-Johnson MJ. Protein C pathway in infants and children. Semin Thromb Hemost 2003;29: Monagle P, Barnes C, Ignjatovic V, Furmedge J, Newall F, Chan A, et al. Developmental haemostasis. Impact for clinical haemostasis laboratories. Thromb Haemost 2006;95: Said JM, Ignjatovic V, Monagle PT, Walker SP, Higgins JR, Brennecke SP. Altered reference ranges for protein C and protein S during early pregnancy: implications for the diagnosis of protein C and protein S deficiency during pregnancy. Thromb Haemost 2010;103:984 8.

72 S66 Bereczky et al.: Protein C and protein S deficiencies 108. Boerger LM, Morris PC, Thurnau GR, Esmon CT, Comp PC. Oral contraceptives and gender affect protein S status. Blood 1987;69: Pabinger I, Kyrle PA, Speiser W, Stoffels U, Jung M, Lechner K. Diagnosis of protein C deficiency in patients on oral anticoagulant treatment: comparison of three different functional protein C assays. Thromb Haemost 1990;63: Demir T, Celkan T, Ahunbay G, Babaoglu A, Besikci R. Venous and intrapericardial thrombosis: secondary to transient protein C deficiency. Pediatr Cardiol 2006;27: Richardson MA, Gupta A, O Brien LA, Berg DT, Gerlitz B, Syed S, et al. Treatment of sepsis-induced acquired protein C deficiency reverses angiotensin-converting enzyme-2 inhibition and decreases pulmonary inflammatory response. J Pharmacol Exp Ther 2008;325: Bay A, Oner AF, Calka O, Sanli F, Akdeniz N, Dogan M. Purpura fulminans secondary to transient protein C deficiency as a complication of chickenpox infection. Pediatr Dermatol 2006;23: Mallet VO, Vallet-Pichard A, Pol S. Human immunodeficiency virus-associated obliterative portopathy underlies unexplained aminotransferase elevations under antiretrovirals. Hepatology 2009;50: Regnault V, Boehlen F, Ozsahin H, Wahl D, de Groot PG, Lecompte T, et al. Anti-protein S antibodies following a varicella infection: detection, characterization and influence on thrombin generation. J Thromb Haemost 2005;3: Ireland H, Thompson E, Lane DA. Gene mutations in 21 unrelated cases of phenotypic heterozygous protein C deficiency and thrombosis. Protein C Study Group. Thromb Haemost 1996;76: Baglin T, Gray E, Greaves M, Hunt BJ, Keeling D, Machin S, et al. Clinical guidelines for testing for heritable thrombophilia. Br J Haematol 2010;149:

73 Clin Chem Lab Med 2010;48(Suppl 1):S67 S by Walter de Gruyter Berlin New York. DOI /CCLM Review Antithrombin deficiency and its laboratory diagnosis László Muszbek 1,2, *, Zsuzsanna Bereczky 1,2, Bettina Kovács 3 and István Komáromi 2 1 Clinical Research Center, University of Debrecen, Medical and Health Science Center, Debrecen, Hungary 2 Thrombosis, Haemostasis and Vascular Biology Research Group of the Hungarian Academy of Sciences, University of Debrecen, Medical and Health Science Center, Debrecen, Hungary 3 Borsod Abaúj Zemplén County and Teaching Hospital, Miskolc, Hungary Abstract Antithrombin (AT) belongs to the serpin family and is a key regulator of the coagulation system. AT inhibits active clotting factors, particularly thrombin and factor Xa; its absence is incompatible with life. This review gives an overview of the protein and gene structure of AT, and attempts to explain how glucosaminoglycans, such as heparin and heparan sulfate accelerate the inhibitory reaction that is accompanied by drastic conformational change. Hypotheses on the regulation of blood coagulation by AT in physiological conditions are discussed. Epidemiology of inherited thrombophilia caused by AT deficiency and its molecular genetic background with genotype-phenotype correlations are summarized. The importance of the classification of AT deficiencies and the phenotypic differences of various subtypes are emphasized. The causes of acquired AT deficiency are also included in the review. Particular attention is devoted to the laboratory diagnosis of AT deficiency. The assay principles of functional first line laboratory tests and tests required for classification are discussed critically, and test results expected in various AT deficiency subtypes are summarized. The reader is provided with a clinically oriented algorithm for the correct diagnosis and classification of AT deficiency, which could be useful in the practice of routine diagnosis of thrombophilia. Clin Chem Lab Med 2010;48:S *Corresponding author: László Muszbek, MD, PhD, Clinical Research Center, University of Debrecen, Medical and Health Science Center, 98, Nagyerdei krt., Debrecen, 4032, Hungary Phone: q , Fax: q , muszbek@med.unideb.hu Received July 30, 2010; accepted September 25, 2010; previously published online November 10, 2010 Keywords: amydolytic assay; antithrombin, antithrombin deficiency; thrombophilia. Introduction The stepwise discovery of antithrombin (AT, SERPINC1) is an exciting story. A detailed account and chronology of the most important events leading to the foundation of our present knowledge about this essential inhibitor of blood coagulation has recently been in an excellent review by Abildgaard (1), recommended for readers interested in this subject. Here, only the issue of nomenclature is mentioned. In the earlier literature, AT used to be termed antithrombin III to distinguish the protein from other proteins with antithrombin activity. Antithrombin I is the thrombin absorbing capacity of fibrin (2) and antithrombin II, more frequently termed heparin cofactor II (SERPIND1) which is another inhibitor of thrombin in plasma (3). Although this is not the only antithrombin present in plasma, we adopted the current nomenclature and use the terms antithrombin and AT throughout the article. AT is a single-chain glycoprotein with a molecular mass of 58,200 Da. The mature protein consists of 432 amino acids with three internal disulfide bonds. As much of the literature and the existing databases use traditional amino acid residue numbering, starting with the N-terminal residue of the mature protein, we used this system in the article. The numbering system recommended by the Human Genome Variation Society (HGVS) starts with the initiator methionine. In the case of AT, HGVS numbering can be calculated by adding 32, corresponding to the 32 amino acid residues of the leader sequence, to the traditional number. AT has two isoforms which differ only in the extent of glycosylation (4, 5). The a isoform is N-glycosylated on four Asn residues (95, 135, 155 and 192), while the b isoform lacks glycosylation on Asn135. The a variant is the major AT isoform (90% 95%) in the circulation, the b isoform represents only 5% 10% of AT in plasma. AT is synthesized in the liver, its half-life in the circulation is approximately 2.4 days. A prominent feature of AT is its high affinity binding to negatively charged glycosaminoglycans (GAGs) such as heparin or heparan sulfate which contain specific pentasaccharide units. Due to the lack of carbohydrate residue on Asn135, the b isoform binds to GAGs with higher affinity. Heparan sulfate in the form of heparan sulfate proteoglycane (HPSG) is present on the surface of vascular endothelium. Thus, a higher portion of the b isoform becomes cleared from the circulation and targets the vessel wall. 2010/447

74 S68 Muszbek et al.: Antithrombin deficiency The structure of antithrombin and structural changes during its interaction with active clotting factors The atomic 3D structure of native AT was resolved approximately 15 years ago (6, 7). Since then, numerous AT structures have been published which have helped in the understanding of how AT exerts its inhibitory function (8 14). AT belongs to the family of serine protease inhibitors (serpins), the largest family of protease inhibitors that consists of over 1500 members (4, 14, 15). These are single chain globular proteins that consist of amino acid residues and show about 30% sequence identity. Serpins share a common tertiary structure; they contain three b-sheets (A-C) and eight to nine a-helices (A-I). A flexible peptide loop, reactive center loop (RCL) containing the reactive site, is exposed on the top of the molecule. RCL contains a sequence which is complementary to the active site of the target protease. All serpins feature significant structural flexibility, which allows dramatic structural changes upon reaction with the protease to be inhibited. These are so-called suicide inhibitors. The target protease cleaves a scissile bond in RCL and then it remains covalently linked to the inhibitor. Antithrombin is a misnomer, the inhibitory effect of AT is not restricted to thrombin. It is a polyvalent serpin that also inhibits activated factor X (FXa) and to a lesser extent a whole series of serine proteases involved in the hemostatic machinery, including FIXa, FXIa, FXIIa, plasmin and kallikrein (4, 5). AT inactivates FVIIa only when it is bound to tissue factor (16 18). AT is a co-called progressive inhibitor; the rate of its reaction with the active coagulation factors is slow, but in the presence of heparin or HPSG, the rate of inhibition is accelerated 500-fold. AT has a typical serpin secondary and tertiary structure. It consists of nine helices and three b-sheets (Figure 1). In the uncleaved form, it can exist in two main conformational states. In the native uncleaved form, the 24-membered RCL with the scissile P1-P1 (Arg393-Ser394) bond is outside the main body of AT (Figure 1A). In the latent conformation, the RCL is inserted into the b-sheet. The latter conformation is thermodynamically more stable than the native form, which is kinetically trapped in a high energy state. AT circulates primarily in this kinetically trapped native form. The X-ray structure of this conformation revealed that the P1 residue (Arg393) points to the surface of the body of AT, and the P14-P15 residues are inserted into b-sheet A, which constrains the RCL and allows contacts between the P1 arginine side chain and the body of AT. Having such a rigid conformation of RCL, AT is a poor inhibitor of FXa or thrombin, and is unable to inhibit FIXa. The binding of pentasaccharide or heparin containing the pentasaccharide unit causes remarkable changes of the conformation of RCL and its close proximity (Figure 2B). The entrapped part of RCL is expulsed from b-sheet A, the end of the third b-strand of b-sheet A moves closer to the fifth b-strand and helix D becomes elongated. The interaction between AT and the pentasaccharide unit takes place in two steps (not shown on Figure 2); an initial weak binding intermediate becomes transformed into a high binding state with 1000-fold higher affinity. The latter conformation is necessary for the effective formation of the Michaelis complex (Figure 2C) between AT and FXa or FIXa, in which Arg393 of AT and the region in its immediate vicinity is recognized by the protease as a substrate loop. The mechanism of Michaelis complex formation between thrombin and AT is somewhat different. In this case, the conformational change induced by the allosteric effect of pentasaccharide is not sufficient, and probably not even required. Thrombin also binds to heparin, and the bridging effect of heparin of 18 saccharide units or longer, which brings thrombin and AT together, is essential for effective interaction (11, 22, 23). After the rate-controlling Michaelis complex formation, the inhibition of active coagulation factors follows the general scheme of serpin action. In the first step of proteolytic reaction, an acyl-enzyme intermediate is formed through an ester bond between Arg393 and the active site serine of the protease. AT undergoes a rapid, drastic and irreversible conformational change where the P14-P3 part of RCL becomes incorporated into b-sheet A as an additional strand (Figure 2D), and AT assumes a cleaved relaxed form. This process is accompanied by a 1000-fold reduction in heparin affinity and by a large-scale conformational change in the acylenzyme complex. The protease which is covalently tethered to Arg393 becomes transported from the top to the bottom of AT, approximately 70 Å away from its original position. Due to the distortion of the active site of the protease, the acyl intermediate becomes stabilized and the second step of proteolytic reaction, the release of the cleaved peptide, cannot take place. In this process the protease structure becomes disrupted and the catalytic triad distorted. Only very slow release of the inactive inhibitor and enzyme can be detected from the AT-protease complex (24). The role of antithrombin in the regulation of coagulation AT serves as a highly important regulator of hemostasis; its absence is incompatible with life (5). The primary actions of AT are the inhibition of thrombin mediated fibrin clot formation and the generation of thrombin by FXa. As mentioned earlier, AT also inhibits activated clotting factors higher up in the intrinsic (FIXa, FXIa, FXIIa) and extrinsic (FVIIa-tissue factor complex) pathways. It also inhibits a series of other non-coagulant effects of these clotting factors, including platelet activation, vascular cell signaling, proliferation, cytokine production, etc. There are two paradoxes concerning the effect of AT and its importance in the regulation of clotting machinery: 1) it is a weak progressive inhibitor of activated clotting factors, 2) it fails to inhibit effectively fibrin-bound thrombin and FXa present in an activation complex on the platelet surface. As discussed in the previous section, interaction with heparin, or its in vivo substitute HSPG, significantly accelerates the inhibitory action of AT and makes it a highly effective inhibitor of thrombin, FXa and other active clotting

75 Muszbek et al.: Antithrombin deficiency S69 Figure 1 Structural elements of antithrombin (AT) in its native (A) and latent (B) state based on the X-ray structure in the RCSB protein data bank (pdb ID: 2b4x). The b-sheets A, B and C are colored yellow, green and purple, respectively. Helical secondary structural elements are shown in orange, except for helix D that is shown in black. Helix D plays a crucial role in heparin pentasaccharide binding. (In the case of serpins, conserved helical structure elements are identified by capital letters.) The P3-P14 portion of the reactive center loop consisting of P1-P17 and P1 -P17 residues is colored dark blue. The Arg393 (P1) residue is shown by a ball-and-stick representation. In the latent state, a substantial portion of the reactive center loop is inserted into b-sheet A as an additional b-strand. The figures were prepared using Chimera software (Resource for Biocomputing, Visualization, and Informatics at the University of California, San Francisco, CA, USA) (19). factors (5). HSPG is widely available on the vascular endothelium and/or in the underlying subendothelial matrix and tissues. Only a minor portion of HSPG contains the 3-Osulfated pentasaccharide unit required for the acceleration of AT activity (25). This active HSPG (A-HSPG) species constitutes 5% of the total HSPG associated with rat microvas- Figure 2 Schematic representation of the mechanism of activation and action of antithrombin (AT) on FXa. (A) The native (circulating) form of AT, (B) the pentasaccharide-activated AT, (C) AT-factor Xa (FXa) complex; only epidermal growth factor 2 (EGF 2) and serine protease domains of FXa are shown, (D) the acyl-enzyme complex in which FXa is covalently linked to AT. For the A, B and C parts of the figure the X-ray coordinates wpdb IDs are 2b4x, 2gd4 (AT only) and 2gd4 (AT-FXa), respectivelyx deposited in the protein data bank (20) were used. The construction of (D) was based on the structure of the a 1 -antitrypsin-elastase acyl-enzyme complex (pdb ID: 2d26) (21). a 1 -antitrypsin was replaced by AT and elastase was substituted by FXa. Next, energy minimization of the constructed complex was performed with Yasara software (Yasara Biosciences GmbH, Vienna, Austria) ( b-sheets and helices are shown in magenta and orange, respectively. Peptide sections that undergo remarkable conformational changes wp1-p15 section of the reactive center loop (RCL), the third b-strand of b-sheet A (s3a) and helix D (hd)x are depicted in green. The Arg393 residue and the pentasaccharide (PeS) are shown by ball-and-stick representations. The figures was prepared using Chimera software (19).

76 S70 Muszbek et al.: Antithrombin deficiency cular endothelial cells. Studies of its distribution in different vessels (26, 27) suggest that A-HSPG present on the surface of endothelial cells provides a basic level of activated AT for anticoagulant activity, while the availability of a much larger pool following endothelial damage and vessel wall injury dramatically increases the anticoagulant potential of AT. The relative importance of progressive AT activity or A- HSPG induced activity in the physiological regulation of blood coagulation is not clear. The fact that homozygous mutations at the heparin binding site, which render AT unable to bind GAGs, are compatible with life (see later for details), as opposed to other homozygous AT deficiencies, suggests the importance of progressive activity. On the other hand, the severe thrombophilia seen in such patients underlines the importance of heparin/hspg induced AT activation. The association of active clotting factors with the surface of activated platelets and with the fibrin clot significantly modifies the inhibitory effect of AT or AT-heparin. When present in the prothombinase complex on platelets or phospholipid surfaces, FXa escapes the inhibition by AT (28 30). Thrombin bound to fibrin or to fibrin degradation products becomes refractive to inhibition by the AT-heparin complex. Fibrin, heparin and thrombin form a ternary complex, in which thrombin exosite 1 and exosite 2 are occupied by fibrin and heparin, respectively (31). This prevents AT-associated heparin from interacting with exosite 2 on fibrinbound thrombin, and the bridging action of heparin cannot operate. It is of interest that FVIIa becomes sensitive to inhibition by AT only when bound to tissue factor (16, 17). The above findings suggest a double role for the AT and AT-A- HSPG complex in the physiological regulation of blood coagulation. It might control low-level thrombin formation that occurs physiologically in the unperturbed circulation by the inhibition of tissue factor-fviia complex, FXa and perhaps other active clotting factors. AT might also exert a scavenger function by neutralizing FXa and thrombin that have escaped from the clot and from the activation complex. The latter mechanism could prevent the propagation of the clot to areas away from the site of vascular injury. In vivo, the AT-protease complex is rapidly eliminated from the circulation by utilizing a common serpin-protease complex clearance pathway. Binding to members of the lowdensity lipoprotein receptor family, primarily to low density lipoprotein-related protein which is an important receptor in the liver, is the main pathway of the elimination of serpinprotease complexes (5, 32). Gene structure of antithrombin The gene for human AT (SERPINC1) is located at the 1q23- q25 position and contains seven exons producing a 1.4-kb messenger RNA (mrna), and six introns (33, 34). All the exon/intron boundaries follow the GT-AG rule. Nine complete and one partial Alu repeats were identified in introns 1, 2, 3B, 4 and 5. There is a highly polymorphic trinucleotide repeat sequence in intron 4 which is useful for haplotype analysis in studies of recurrent mutations and for linkage analysis in families with thrombosis (35). Primer extension analysis has mapped the AT transcriptional start site in liver cells to a position 72 bp upstream of the ATG translation initiator codon (36). A leader sequence of 32 amino acids is encoded by exon 1 and the 5 end of exon 2 (37). The heparin binding site of AT is encoded by exon 2 and exon 3a. The reactive site, located in the carboxy-terminal part of the protein, is encoded by exon 6. Epidemiology of antithrombin deficiency The prevalence of inherited AT deficiency in the general population is estimated to be between 1:2000 and 1:3000 (38). Most of the genetic defects result in type II (qualitative) deficiencies (39). The prevalence of AT deficiency in patients with venous thromboembolism (VTE) is much higher, between 1:20 and 1:200 (40). According to an Italian study of symptomatic patients and relatives, type I mutations (quantitative deficiencies) are more frequent than type II variants (41). In unselected patients with a history of VTE, the frequency of AT deficiency is 0.5% 1.1% (40, 42). In a cumulated analysis of 1705 selected patients with VTE, the frequency of AT deficiency was 2.4% (43). During a mean follow-up time of 2.3 years the incidence of venous thrombosis was high; being 12% in individuals with hereditary AT deficiency in a small Italian cohort (44). For comparison, the incidence of thrombosis in protein C (PC) and protein S (PS) deficiency was 2.8% and 3.3%, respectively. In the large prospective EPCOT study (European Prospective Cohort on Thrombophilia), the risk of first VTE in asymptomatic AT, PC or PS deficient individuals and in individuals with Factor V Leiden mutation was analyzed (ns575). During the 5.7 year of follow-up, 4.5% of these individuals developed VTE, the annual incidence of first VTE was the highest in those with AT deficiency (1.7%/ year) (45). Based on the prevalence data in the general population and in VTE patients, the relative risk of VTE in patients with AT deficiency was estimated to be approximately fold (46). Since then, prospective and casecontrol studies have calculated the same magnitude of VTE risk conferred by AT deficiency in different ethnical groups (47 49). Based on the results of these epidemiological studies, it can be concluded that the risk of VTE conferred by hereditary AT deficiency is the highest among inherited thrombophilias. However, the risk of VTE seems to vary according to the subtypes of AT deficiency (see later). AT deficiency also represents an increased risk for development of PE in deep venous thrombosis (DVT), and an increased risk for recurrence of VTE. In an Italian study of patients with proximal DVT, the risk of pulmonary embolism (PE) was 2.4-fold (95% CI: ) in AT deficient patients compared to individuals who developed DVT without inherited thrombophilia (50). In AT deficiency, the annual incidence of recurrent VTE was found to be 10% (95% CI: 6.1% 15.4%) in a recently released Dutch study (51). In

77 Muszbek et al.: Antithrombin deficiency S71 an Italian cohort, the adjusted hazard ratio for the recurrence of VTE was 1.9 (95% CI: ) (52). Molecular genetic background of antithrombin deficiency, genotype-phenotype correlations The first report on AT deficiency was described by Egeberg in 1965 (53). The first functional AT defect, AT Budapest, was reported by Sas et al. in 1974 (54). Since then, a high number of deficient patients have been identified, and the molecular genetic background was clarified in a significant number of cases. According to the recommendations of the International Society on Thrombosis and Haemostasis, AT deficiency is classified as type I (quantitative) and type II (qualitative) deficiency (55). In type I deficiency, AT activity and the antigen concentration are equally decreased, suggesting defective synthesis or secretion of the protein. In type II deficiency, the defect may involve the reactive site (type II RS), the heparin-binding site (type II HBS) or it can exert a pleiotropic effect (type II PE) (56). The inheritance of AT deficiency, in general, is autosomal dominant. However, in the case of type II HBS deficiency, it often shows incomplete penetrance or an autosomal recessive pattern. The majority of AT deficient patients are heterozygous for the defect with typical AT activity values approximately 50%. Homozygosity is incompatible with life, with the exception of type II HBS variant (described later). The molecular genetic background of AT deficiency is heterogeneous. The mutations are best summarized in the Antithrombin Mutation Database ( mentofmedicine/experimentalmedicine/haematology/coag/- antithrombin/) and in the database of human gene mutation data (HGMD) ( (Figure 3). Almost 50% of the 215 different mutations that have been reported in the HGMD are missense mutations. Small deletions and insertions are also common, contributing 20% and 10%, respectively. Non-sense mutations and splicing site mutations represent 8% and 5% of all reported causative sequence variants, respectively. Whole or partial gene deletions are relatively frequent (5%), while complex rearrangements are rare. Type I AT deficiencies are most commonly caused by insertions or deletions leading to frameshift and premature stop codon, or less commonly by non-sense mutations. These mutations obviously explain the type I phenotype, primarily a result of unstable mrna transcripts and/or the presence of truncated proteins. Large gene segment deletions also lead to type I deficiency. By screening mutation negative AT deficient cases using the multiplex ligation-dependent probe amplification (MLPA) technique, several gross deletions were identified. The breakpoints are often located within Alu repeat elements (57). Amino acid changes caused by single nucleotide substitutions within the coding region of SER- PINC1 may also lead to type I deficiency. In this case, the absence of mutant protein in the circulation is due to misfolding or a secretion defect (4). The type II AT deficiencies are most commonly caused by missense mutations. Among the mutations known to involve the reactive site domain, two regions are preferred: the hinge region (most frequently residues Ala382 and Ala384) and around the reactive domain at residues Gly392 (AT Stockholm), Arg393 and Ser394 (58). Most of the missense mutations leading to type II HBS deficiency affect residues Pro41 (AT Basel), Arg47 (AT Padua I), Leu99 (AT Budapest 3) and Arg129. Practically all patients with AT Budapest 3 (p.leu99phe) mutation described to date were of South Eastern European origin, which may suggest a founder effect (35, 59 61). Type II PE deficiency is caused by mutations involving residues 402, and 429. This region is responsible for both the structural and functional integrity of AT. These mutations lead to impaired function of the reactive site and also to reduced secretion (58). A new pleiotropic mutant (AT Murcia, p.k241e) has been described recently in which altered glycosylation of the molecule led to impaired heparin binding and thrombin inhibition (62). Some of the missense mutations occurred in the mobile regions of AT, mainly at the hinges of the reactive center loop, or in the region involved in the shutter-like opening of the main b-sheet of the molecule. The latter is required for insertion of the reactive loop into the b-sheet. Even change in a single amino acid in these sensitive regions can lead to conformational changes, loss of stability that facilitates the formation of intermolecular linkages and lead to the formation of oligomers or transformation to the latent conformation (4). The secretion of these variants is also impaired which results in a circulating deficiency. Homozygous type I AT deficiency is not compatible with life and heterozygous patients usually suffer severe thrombosis at a young age. The same stands for type II RS and type II PE deficiencies. However, there is at least one notable exception. The heterozygous p.ala384ser mutation (AT Cambridge II) causes type II RS deficiency with a mild phenotype, and this mutation can also exist in homozygous form (63, 64). Type II HBS deficiency confers a lower risk of thrombosis compared with the other subtypes (65, 66). Homozygous type II HBS patients usually survive, thrombosis may develop even earlier (frequently in childhood) than in patients with heterozygous type I or other type II deficiencies. Symptoms of AT deficiency are DVT and/or PE which are often recurrent. DVT not infrequently develops at unusual sites, such as in the proximal extremities, and in mesenteric, renal, portal, retinal and cerebral veins (67 70). Intracardial atrial thrombosis has also been reported (71). The risk of thrombosis conferred by AT deficiency to pregnant women is significantly greater than in other deficiencies. The estimated risk is 1:2.8 for women with type I deficiency, which is approximately 350-times higher than the risk conferred by pregnancy alone (72, 73). In addition to venous thrombosis, occasionally, arterial thrombosis has also been reported in patients with AT deficiency (74, 75). AT Cambridge (p.a384s) mutation increased the risk of myocardial infarction 5.66-fold as reported by a Spanish study that enrolled 1224 patients and

78 S72 Muszbek et al.: Antithrombin deficiency Figure 3 Distribution of causative mutations in the SERPINC1 gene according to the database of human gene mutation data (HGMD) that have been published to date ( Nucleotide numbering is given relative to the first nucleic acid of the initiator ATG codon. Amino acids are numbered according to the mature protein, where the first methionine is numbered controls (76). In contrast, in a large cohort of relatives of VTE patients with PC, PS or AT deficiency (ns468), the risk of arterial thrombosis in those -55 year of age was increased only in PC and PS deficient patients, while AT deficiency was not associated with an increased risk (hazard ratio 1.1, 95% CI: ) (77). According to a recent meta-analysis of studies involving children with arterial ischemic stroke and cerebral venous sinus thrombosis, the summary OR for the risk of arterial ischemic stroke in children having AT deficiency was 3.29 (95% CI: ), while the OR for the risk of cerebral venous sinus thrombosis was (95% CI: ) (78). These findings do not support a significant contribution of AT deficiency to the risk of atherothrombotic events. Acquired antithrombin deficiency In healthy full-term newborns, the concentration of AT is in the range of 51% 75% of adult average values (79). Due to severe immaturity of the liver, in preterm infants AT concentrations can be much lower than in full terms infants (80). AT concentrations reach adult ranges by the age of 1 year (81). Production of AT is reduced in liver disease with impaired hepatic function. In patients with nephrotic syndrome or other diseases associated with renal or enteral protein loss, the low AT concentration is due to increased elimination. Low concentrations of AT as a result of consumption are found in patients with sepsis, disseminated intravascular coagulation, large thrombus, thrombotic micro-

79 Muszbek et al.: Antithrombin deficiency S73 angiopathy, acute hemolytic transfusion reactions and malignancies (82). There are two notable examples of druginduced AT deficiency. Long-term therapy with unfractionated heparin is a common cause of moderate AT consumption, which is probably the result of greatly enhanced formation of thrombin-at complex in plasma. Therapy with L-asparaginase leads to intracellular retention of AT within the endoplasmic reticulum, perhaps due to interference with folding of the molecule and with glycosylation of the protein (83, 84). Interestingly, co-administration of dexaamethasone with L-asparaginase increased the concentration of AT and reduced the risk of thrombosis. The effect of dexamethasone is possibly due to induced expression of heat shock proteins and endoplasmic reticulum-associated chaperons which prevent the conformational effect of L-asparaginase (85). Laboratory diagnosis of antithrombin deficiency A first-line test for the diagnosis of AT deficiency should detect all deficiencies, i.e., AT deficiencies due to decreased AT concentration as well as to a defective molecule. Therefore, the first line test should be a functional assay. The original clotting methods where the inhibition of thrombin by diluted native serum or defibrinated plasma was measured by fibrinogen clotting are impractical and inaccurate and not in use any longer. With the modern chromogenic (amidolytic) assays, the inhibition of thrombin or FXa activity by AT is measured using thrombin/fxa specific tri, or tetra-peptide substrates which show sequential similarity to the P1-P3 or P1-P4 sequences of the natural substrates of these enzymes (Figure 4) (86, 87). The peptides conform to the active site of the respective active clotting factor and a para-nitroanaline (pna) group is attached to their C-terminal end. Thrombin or FXa rapidly release the pna group from their peptide substrate. Free pna, as opposed to the peptide-bound form, has strong light absorption at 405 nm and its release can be easily monitored spectrophotometrically. The assays can be performed in the presence of heparin (heparin cofactor activity) or without heparin (progressive activity). In the former assays, the inhibition of active clotting factors is very quick, while in the latter cases more time is required for ATII to exert its inhibitory action. As only the heparin cofactor activity is decreased in all subtypes of AT deficiency (Table 1), the assay measuring this activity is the generally accepted first line test for the diagnosis of deficiency. Unfortunately, as external quality control exercises reveal, the improper practice of using only an antigenic AT assay, which detects -50% of AT deficiencies, still exists in a few laboratories. Figure 4 demonstrates the assay principle of amidolytic AT assays. Heparin binds to AT making it highly reactive with thrombin and FXa (activated AT; AT*). Thrombin or FXa is added in excess of AT and a part of it becomes rapidly complexed with heparin-at*, in the complex AT activity is abrogated. The extent of thrombin/fxa inhibition depends on plasma AT activity, and the residual free thrombin or FXa is inversely related to AT activity. The amount of free thrombin or FXa is measured using a chromogenic substrate described above. The increase in absorbance at 405 nm can be measured using a kinetic or end-point method (the former is preferred), and the change of absorbance is converted to AT activity using a calibration curve. Reference plasma of known AT activity is used to construct the calibration curve. A WHO international standard (2nd International Standard Antithrombin, Plasma, NIBSC code: 93/768) with an assigned potency of 0.85 International Units (IU) is available from the National Institute for Biological Standards and Control (NIBSC; Potters Bar, UK). This international plasma standard should be used by companies for the calibration of their reference plasma and this information should be stated on the application sheet. The chromogenic heparin cofactor AT assay has good reproducibility. Laboratories equipped with automated laboratory analyzers or coagulometer should aim for within-batch precision CVF2%, and for within-laboratory reproducibility (total error) CVF5%. According to a previous report based on the quality assessment program for thrombophilia screening by the ECAT Foundation which included 136 laboratories during the time period , the median long-term within-laboratory analytical CV was 7.6% with a 95% CI of (88). These values suggest that for many laboratories, there is much work needed to improve the quality Table 1 Laboratory diagnosis and classification of antithrombin (AT) deficiencies. Figure 4 Measurement principle of chromogenic antithrombin assays. Both anti-thrombin and anti-fxa heparin cofactor assays are demonstrated. AT, antithrombin; AT*, antithrombin activated by heparin; FXa, activated factor X; R, the peptide part of chomogenic thrombin or FXa substrates; pna, para-nitro aniline. In the last line, the oligopeptide components of a thrombin (thr) substrate (S-2238) and a FXa substrate (S-2772) are shown. Subtypes of AT Heparin cofactor Progressive AT antigen deficiencies AT assay AT assay assay Type I x x x Type II RS x x n Type II HBS x n n Type II PE x x n or subnormal RS, reactive site; HBS, heparin binding site; PE, pleiotrop; n, normal.

80 S74 Muszbek et al.: Antithrombin deficiency performance of AT assays. Joining an international accredited external quality assessment program is highly recommended for laboratories routinely performing AT measurements. Human thrombin was used in previous thrombin inhibition assays. Human thrombin also reacted with heparin cofactor II and made the assay relatively insensitive for the detection of AT deficiency (89, 90). In most commercial kits, human thrombin has been replaced by bovine thrombin which shows minimal reaction with heparin cofactor II. FXa does not react with heparin cofactor II at all. Heparin cofactor AT assays based on bovine thrombin and FXa inhibition seem to function equally well, the sensitivity of both assays is close to 100%. As the reactive site of thrombin and FXa differs somewhat, one would expect that AT deficiencies caused by certain mutations around the reactive site are detected by the two types of assays with different sensitivity. Indeed, the Ala384Ser mutation (AT Cambridge II) which is a relatively prevalent variant in the general population, is not detected by the anti-fxa assay, but anti-thrombin activity is mildly, but significantly, reduced (63, 64). However, this mutation causes only a relatively mild thrombophilia, and even elderly homozygous individuals might lack the history of VTE. In contrast, according to our experience with a high number of AT Budapest 3 mutants, the anti-xa assay is significantly more sensitive in detecting this type II HBS deficiency compared with the assay based on thrombin inhibition. Progressive AT assays are based on the same principle as heparin cofactor assays, but are performed in the absence of heparin on less diluted plasma samples. In addition, the incubation time is prolonged significantly. Unfortunately, the application sheets for the commercial AT activity assays do not provide a description on how to use the test kit as a progressive assay, the appropriate conditions need to be established by the user. These conditions may vary among commercial kits, and the user needs to experiment. As a starting point, 10-fold diluted plasma and a 15-min incubation time is recommended. Although this test is essential for the diagnosis of type II HBS deficiency (Table 1), for the reasons mentioned above, the progressive assay is very much under utilized in the diagnosis and classification of AT deficiencies (91). As discussed earlier, differentiation between heterozygous type II HBS deficiency causing mild thrombophilia and other type of functional defects causing a severe phenotype is of clinical relevance. In certain, clinical set-ups this might influence the decision concerning anticoagulant therapy. Distinguishing between homozygous type II HBS deficient patients who have a very severe phenotype, and heterozygotes is also of clinical relevance. Measurement of AT antigen concentrations is required for the classification of AT deficiencies. Traditional electroimmunodiffusion and radial-immunodiffusion techniques are too time consuming, imprecise and their use is no longer recommended. At present, latex-enhanced immuno nephelometry is the most frequently used method for measurement of AT antigen concentrations (92), and commercial kits for this purpose are available. It is rather surprising that no reference interval determined according to guideline (C28-A3) from the Clinical and Laboratory Standards Institute (CLSI; Wayne, PA, USA) is available for AT activity and antigen. A number of different normal ranges, varying within a narrow interval have been reported in the literature and are available in manufacturer s application sheets. Accepting 80% of the average normal (0.8 IU/mL) as the lower limit of reference interval for AT activity seems to be an acceptable compromise, and most laboratories use this value. The upper limit of the reference interval does not have any clinical relevance. AT antigen concentrations can be expressed as mass concentration (mg/l), although there are a rather wide variety of normal ranges for mass concentration. In addition, values for mass concentration are difficult to compare to AT activity values. For this reason most laboratories use the same principle and the same reference interval for AT antigen as for AT activity. It should be noted that by definition, 2.5% of the values obtained with normal, non-deficient samples are below the lower limit of the reference interval, and values between 70% and 80% should be interpreted with extreme caution. Platelet poor citrated plasma is used for both activity and antigen measurements and can be stored at 208C for up to 4 months. Measurement of AT activity and concentration is not recommended within 3 months of an acute event. During this period, if values are within the reference interval, the exclusion of AT deficiency is possible, but the diagnosis of AT deficiency cannot be confirmed. In a number of cases, AT determination is requested for patients who are on anticoagulant therapy. Oral anticoagulant therapy with vitamin K antagonists, such as warfarin or acenocoumarol might increase the level of AT (93 98), while administration of unfractionated heparin decreases the concentration of AT (58, 94, 97, 99). Low molecular weight heparins do not have such an effect (100). For these reasons, we do not recommend diagnosing AT deficiency in patients who are undergoing unfractionated heparin therapy. Also, we do not recommend attempting to exclude AT deficiency during oral anticoagulant therapy. In our experience switching from oral anticoagulant therapy to low molecular heparin for 10 days prior to blood collection is a good compromise which allows the measurement of valid AT values. The algorithm used in the authors laboratory for the diagnosis and classification of AT deficiency is demonstrated in Figure 5. If heparin cofactor AT activity is -80% we carefully look for and exclude acquired AT deficiencies, such as liver disease, renal or enteral protein loss, consumption coagulopathy, unfractionated heparin treatment or therapy with l-asparaginase. To establish the diagnosis, repeated tests are required from different blood samples collected from the same person. If feasible, we recommend a time interval of at least 3 weeks between the two blood collections. If the heparin cofactor activity is repeatedly equal to or -70% and acquired causes have been excluded, the laboratory diagnosis of inherited AT deficiency can be established. Between 70% and 80% of AT activity, it is highly recommended to confirm the diagnosis by molecular genetic testing. Once AT deficiency is diagnosed, the next step is its classification, which occurs in two steps. Plasma AT antigen concentrations are measured to differentiate between type I and type II deficiency. Decreased

81 Muszbek et al.: Antithrombin deficiency S75 Figure 5 Laboratory diagnosis and classification of antithrombin (AT) deficiency. Assays are shown in rectangles, diagnoses in italics are underlined. As stated in the text, we do not recommend diagnosing AT deficiency in patients receiving therapy with unfractionated heparin, and do not recommend excluding AT deficiency during oral anticoagulant therapy. AT antigen implies type I deficiency, while AT antigen in the normal range indicates type II deficiency. Finally, it is clinically important to distinguish II HBS subtype from other type II variants by performing a progressive activity assay. As opposed to other type II subtypes, HBS variants have normal progressive activity. Table 1 summarizes the results of diagnostic and classification tests in different subtypes of AT deficiency. Concluding remarks AT is a slow progressive inhibitor of active clotting factors, particularly thrombin and factor Xa (FXa). Glycosaminoglycans with a 3-O-sulfated pentasaccharide unit, like heparin or heparan sulfate, bind to AT with high affinity and greatly accelerate the reaction with active clotting factors. Even in the presence of heparin/heparan sulfate, AT poorly inhibits FXa in activation complex and fibrin-bound thrombin. It might control low level thrombin formation that occurs physiologically, and could also exert a scavenger function by neutralizing FXa and thrombin that have escaped from the clot and from the activation complex. In general, inherited AT deficiency causes severe thrombophilia. However, the phenotypic appearance varies with different subtypes. Homozygous type I AT deficiency caused by decreased synthesis or secretion is incompatible with life. In the heterozygous form, it is frequently accompanied by DVT, not infrequently of unusual localization, or PE after the second decade of life. Functional defects (type II AT deficiencies) caused by missense mutations are classified according to the site that is affected by the mutation as reactive site (RS), heparin binding site (HBS) AT deficiencies or multiple site defect caused by mutations with pleiotropic effects (PE). Type II HBS subtype is less severe than other AT deficiencies. In the heterozygous form it presents with only mild thrombophilia, and homozygotes also survive, although usually with very early thrombotic complications. The diagnosis of inherited AT deficiency is important for establishing the risk of recurrent thrombotic events. The diagnosis might influence the clinical decision concerning the duration of anticoagulant therapy. The diagnosis is established by laboratory tests, although the exclusion of acquired deficiency requires careful clinical attention. The first line test is a functional chromogenic heparin cofactor assay, which measures the inhibition of thrombin or the inhibition of FXa in the presence of heparin. Both anti-thrombin and anti-fxa assays perform well and, with very few exceptions, detect all subtypes of AT deficiency. The reference interval is quite narrow and in most cases the range of 80% 120% of average normal is accepted. Measurement of AT antigen concentrations allows differentiation between type I and type II AT deficiencies, while progressive AT activity assays are required to make the clinically important distinction between type II HBS and other type II subtypes. Increased utilization of a progressive activity assay is desirable. The diagnosis might be confirmed by molecular genetic testing, which is important in the case of activities in the range of 70% 80%. Acknowledgements This work was supported by a grant from the Hungarian National Research Fund (OTKA K78386). Conflict of interest statement Author s conflict of interest disclosure: The authors stated that there are no conflicts of interest regarding the publication of this article. Research funding: None declared. Employment or leadership: None declared. Honorarium: None declared. References 1. Abildgaard U. Antithrombin early prophecies and present challenges. Thromb Haemost 2007;98: Mosesson MW. Update on antithrombin I (fibrin). Thromb Haemost 2007;98: Rau JC, Beaulieu LM, Huntington JA, Church FC. Serpins in thrombosis, hemostasis and fibrinolysis. J Thromb Haemost 2007;5 Suppl 1: Hernandez-Espinosa D, Ordonez A, Vicente V, Corral J. Factors with conformational effects on haemostatic serpins: implications in thrombosis. Thromb Haemost 2007;98:

82 S76 Muszbek et al.: Antithrombin deficiency 5. Bock SC. Antithrombin III and heparin cofactor II. In: Colman RW, Clowes AW, Goldhaber SZ, Marder VJ, George JN, editors. Hemostasis and Thrombosis, Vol. Philadelphia: Lippincott, 2006: Carrell RW, Stein PE, Fermi G, Wardell MR. Biological implications of a 3-A structure of dimeric antithrombin. Structure 1994;2: Schreuder HA, de Boer B, Dijkema R, Mulders J, Theunissen HJ, Grootenhuis PD, et al. The intact and cleaved human antithrombin III complex as a model for serpin-proteinase interactions. Nat Struct Biol 1994;1: Whisstock JC, Pike RN, Jin L, Skinner R, Pei XY, Carrell RW, et al. Conformational changes in serpins: II. The mechanism of activation of antithrombin by heparindagger. J Mol Biol 2000; 301: Huntington JA, Gettins PG. Conformational conversion of antithrombin to a fully activated substrate of factor Xa without need for heparin. Biochemistry 1998;37: Whisstock JC, Bottomley SP. Molecular gymnastics: serpin structure, folding and misfolding. Curr Opin Struct Biol 2006;16: Johnson DJ, Li W, Adams TE, Huntington JA. Antithrombin- S195A factor Xa-heparin structure reveals the allosteric mechanism of antithrombin activation. EMBO J 2006;25: Langdown J, Belzar KJ, Savory WJ, Baglin TP, Huntington JA. The critical role of hinge-region expulsion in the induced-fit heparin binding mechanism of antithrombin. J Mol Biol 2009; 386: Johnson DJ, Langdown J, Huntington JA. Molecular basis of factor IXa recognition by heparin-activated antithrombin revealed by a 1.7-A structure of the ternary complex. Proc Natl Acad Sci USA 2010;107: Huntington JA. Shape-shifting serpins advantages of a mobile mechanism. Trends Biochem Sci 2006;31: Silverman GA, Bird PI, Carrell RW, Church FC, Coughlin PB, Gettins PG, et al. The serpins are an expanding superfamily of structurally similar but functionally diverse proteins. Evolution, mechanism of inhibition, novel functions, and a revised nomenclature. J Biol Chem 2001;276: Rao LV, Rapaport SI, Hoang AD. Binding of factor VIIa to tissue factor permits rapid antithrombin III/heparin inhibition of factor VIIa. Blood 1993;81: Lawson JH, Butenas S, Ribarik N, Mann KG. Complexdependent inhibition of factor VIIa by antithrombin III and heparin. J Biol Chem 1993;268: Broze GJ Jr, Likert K, Higuchi D. Inhibition of factor VIIa/ tissue factor by antithrombin III and tissue factor pathway inhibitor. Blood 1993;82: Pettersen EF, Goddard TD, Huang CC, Couch GS, Greenblatt DM, Meng EC, et al. UCSF Chimera a visualization system for exploratory research and analysis. J Comput Chem 2004; 25: Berman HM, Westbrook J, Feng Z, Gilliland G, Bhat TN, Weissig H, et al. The Protein Data Bank ( Nucleic Acids Research 2000;28: Dementiev A, Dobo J, Gettins PG. Active site distortion is sufficient for proteinase inhibition by serpins: structure of the covalent complex of alpha1-proteinase inhibitor with porcine pancreatic elastase. J Biol Chem 2006;281: Dementiev A, Petitou M, Herbert JM, Gettins PG. The ternary complex of antithrombin-anhydrothrombin-heparin reveals the basis of inhibitor specificity. Nat Struct Mol Biol 2004;11: Li W, Johnson DJ, Esmon CT, Huntington JA. Structure of the antithrombin-thrombin-heparin ternary complex reveals the antithrombotic mechanism of heparin. Nat Struct Mol Biol 2004;11: Bjork I, Jackson CM, Jornvall H, Lavine KK, Nordling K, Salsgiver WJ. The active site of antithrombin. Release of the same proteolytically cleaved form of the inhibitor from complexes with factor IXa, factor Xa, and thrombin. J Biol Chem 1982;257: Kojima T, Leone CW, Marchildon GA, Marcum JA, Rosenberg RD. Isolation and characterization of heparan sulfate proteoglycans produced by cloned rat microvascular endothelial cells. J Biol Chem 1992;267: de Agostini AI, Watkins SC, Slayter HS, Youssoufian H, Rosenberg RD. Localization of anticoagulantly active heparan sulfate proteoglycans in vascular endothelium: antithrombin binding on cultured endothelial cells and perfused rat aorta. J Cell Biol 1990;111: Faulk WP, Labarrere CA. Modulation of vascular antithrombin III in human cardiac allografts. Haemostasis 1993;23 Suppl 1: Marciniak E. Factor-Xa inactivation by antithrombin. 3. Evidence for biological stabilization of factor Xa by factor V-phospholipid complex. Br J Haematol 1973;24: Teitel JM, Rosenberg RD. Protection of factor Xa from neutralization by the heparin-antithrombin complex. J Clin Invest 1983;71: Lindhout T, Baruch D, Schoen P, Franssen J, Hemker HC. Thrombin generation and inactivation in the presence of antithrombin III and heparin. Biochemistry 1986;25: Becker DL, Fredenburgh JC, Stafford AR, Weitz JI. Exosites 1 and 2 are essential for protection of fibrin-bound thrombin from heparin-catalyzed inhibition by antithrombin and heparin cofactor II. J Biol Chem 1999;274: Kounnas MZ, Church FC, Argraves WS, Strickland DK. Cellular internalization and degradation of antithrombin IIIthrombin, heparin cofactor II-thrombin, and alpha 1-antitrypsin-trypsin complexes is mediated by the low density lipoprotein receptor-related protein. J Biol Chem 1996;271: Olds RJ, Lane DA, Chowdhury V, De Stefano V, Leone G, Thein SL. Complete nucleotide sequence of the antithrombin gene: evidence for homologous recombination causing thrombophilia. Biochemistry 1993;32: Bock SC, Harris JF, Balazs I, Trent JM. Assignment of the human antithrombin III structural gene to chromosome 1q Cytogenet Cell Genet 1985;39: Olds RJ, Lane DA, Chowdhury V, Sas G, Pabinger I, Auberger K, et al. (ATT) Trinucleotide repeats in the antithrombin gene and their use in determining the origin of repeated mutations. Hum Mutat 1994;4: Prochownik EV, Orkin SH. In vivo transcription of a human antithrombin III minigene. J Biol Chem 1984;259: Aiach M, Gandrille S, Emmerich J. A review of mutations causing deficiencies of antithrombin, protein C and protein S. Thromb Haemost 1995;74: Tait RC, Walker ID, Perry DJ, Islam SI, Daly ME, McCall F, et al. Prevalence of antithrombin deficiency in the healthy population. Br J Haematol 1994;87: McColl M, Tait RC, Walker ID, Perry DJ, McCall F, Conkie JA. Low thrombosis rate seen in blood donors and their relatives with inherited deficiencies of antithrombin and protein C: correlation with type of defect, family history, and absence of

83 Muszbek et al.: Antithrombin deficiency S77 the factor V Leiden mutation. Blood Coagul Fibrinolysis 1996;7: Mateo J, Oliver A, Borrell M, Sala N, Fontcuberta J. Laboratory evaluation and clinical characteristics of 2,132 consecutive unselected patients with venous thromboembolism results of the Spanish Multicentric Study on Thrombophilia (EMET- Study). Thromb Haemost 1997;77: Martinelli I, Mannucci PM, De Stefano V, Taioli E, Rossi V, Crosti F, et al. Different risks of thrombosis in four coagulation defects associated with inherited thrombophilia: a study of 150 families. Blood 1998;92: Heijboer H, Brandjes DP, Buller HR, Sturk A, ten Cate JW. Deficiencies of coagulation-inhibiting and fibrinolytic proteins in outpatients with deep-vein thrombosis. N Engl J Med 1990;323: De Stefano V, Finazzi G, Mannucci PM. Inherited thrombophilia: pathogenesis, clinical syndromes, and management. Blood 1996;87: Finazzi G, Barbui T. Different incidence of venous thrombosis in patients with inherited deficiencies of antithrombin III, protein C and protein S. Thromb Haemost 1994;71: Vossen CY, Conard J, Fontcuberta J, Makris M, van der Meer FJ, Pabinger I, et al. Risk of a first venous thrombotic event in carriers of a familial thrombophilic defect. The European Prospective Cohort on Thrombophilia (EPCOT). J Thromb Haemost 2005;3: Rosendaal FR. Risk factors for venous thrombotic disease. Thromb Haemost 1999;82: Sakata T, Okamoto A, Mannami T, Matsuo H, Miyata T. Protein C and antithrombin deficiency are important risk factors for deep vein thrombosis in Japanese. J Thromb Haemost 2004;2: Lijfering WM, Brouwer JL, Veeger NJ, Bank I, Coppens M, Middeldorp S, et al. Selective testing for thrombophilia in patients with first venous thrombosis: results from a retrospective family cohort study on absolute thrombotic risk for currently known thrombophilic defects in 2479 relatives. Blood 2009;113: Brouwer JL, Veeger NJ, Kluin-Nelemans HC, van der Meer J. The pathogenesis of venous thromboembolism: evidence for multiple interrelated causes. Ann Intern Med 2006;145: Rossi E, Za T, Ciminello A, Leone G, De Stefano V. The risk of symptomatic pulmonary embolism due to proximal deep venous thrombosis differs in patients with different types of inherited thrombophilia. Thromb Haemost 2008;99: Brouwer JL, Lijfering WM, Ten Kate MK, Kluin-Nelemans HC, Veeger NJ, van der Meer J. High long-term absolute risk of recurrent venous thromboembolism in patients with hereditary deficiencies of protein S, protein C or antithrombin. Thromb Haemost 2009;101: De Stefano V, Simioni P, Rossi E, Tormene D, Za T, Pagnan A, et al. The risk of recurrent venous thromboembolism in patients with inherited deficiency of natural anticoagulants antithrombin, protein C and protein S. Haematologica 2006;91: Egeberg O. Inherited antithrombin deficiency causing thrombophilia. Thromb Diath Haemorrh 1965;13: Sas G, Blasko G, Banhegyi D, Jako J, Palos LA. Abnormal antithrombin III (antithrombin III Budapest ) as a cause of a familial thrombophilia. Thromb Diath Haemorrh 1974;32: Lane DA, Bayston T, Olds RJ, Fitches AC, Cooper DN, Millar DS, et al. Antithrombin mutation database: 2nd (1997) update. For the Plasma Coagulation Inhibitors Subcommittee of the Scientific and Standardization Committee of the International Society on Thrombosis and Haemostasis. Thromb Haemost 1997;77: Lane DA, Olds RJ, Conard J, Boisclair M, Bock SC, Hultin M, et al. Pleiotropic effects of antithrombin strand 1C substitution mutations. J Clin Invest 1992;90: Picard V, Chen JM, Tardy B, Aillaud MF, Boiteux-Vergnes C, Dreyfus M, et al. Detection and characterisation of large SER- PINC1 deletions in type I inherited antithrombin deficiency. Hum Genet 2010;127: Patnaik MM, Moll S. Inherited antithrombin deficiency: a review. Haemophilia 2008;14: Kuhle S, Lane DA, Jochmanns K, Male C, Quehenberger P, Lechner K, et al. Homozygous antithrombin deficiency type II (99 Leu to Phe mutation) and childhood thromboembolism. Thromb Haemost 2001;86: Olds RJ, Lane DA, Boisclair M, Sas G, Bock SC, Thein SL. Antithrombin Budapest 3. An antithrombin variant with reduced heparin affinity resulting from the substitution L99F. FEBS Lett 1992;300: Sas G, Peto I, Banhegyi D, Blasko G, Domjan G. Heterogeneity of the classical antithrombin III deficiency. Thromb Haemost 1980;43: Martinez-Martinez I, Ordonez A, Navarro-Fernandez J, Perez- Lara A, Gutierrez-Gallego R, Giraldo R, et al. Antithrombin Murcia (K241E) causing antithrombin deficiency: a role for altered glycosylation. Haematologica 2010;95: Perry DJ, Daly ME, Tait RC, Walker ID, Brown K, Beauchamp NJ, et al. Antithrombin cambridge II (Ala384Ser): clinical, functional and haplotype analysis of 18 families. Thromb Haemost 1998;79: Corral J, Hernandez-Espinosa D, Soria JM, Gonzalez-Conejero R, Ordonez A, Gonzalez-Porras JR, et al. Antithrombin Cambridge II (A384S): an underestimated genetic risk factor for venous thrombosis. Blood 2007;109: Finazzi G, Caccia R, Barbui T. Different prevalence of thromboembolism in the subtypes of congenital antithrombin III deficiency: review of 404 cases. Thromb Haemost 1987;58: Girolami A, Lazzaro AR, Simioni P. The relationship between defective heparin cofactor activities and thrombotic phenomena in AT III abnormalities. Thromb Haemost 1988;59: Harmanci O, Ersoy O, Gurgey A, Buyukasik Y, Gedikoglu G, Balkanci F, et al. The etiologic distribution of thrombophilic factors in chronic portal vein thrombosis. J Clin Gastroenterol 2007;41: Ates O. The deficiencies of protein C, protein S and antithrombin III in patients with retinal vein occlusion: a Turkish sample. Clin Lab Haematol 2006;28: Calcaterra D, Martin JT, Ferneini AM, De Natale RW. Acute mesenteric and aortic thrombosis associated with antithrombin deficiency: a rare occurrence. Ann Vasc Surg 2010;24:415 e Shibahara K, Tatsuta K, Orita H, Yonemura T, Kohno H. Superior mesenteric and portal vein thrombosis caused by congenital antithrombin III deficiency: report of a case. Surg Today 2007; 37: Santangeli P, Sestito A. Acute left atrial thrombosis during anticoagulant therapy in a patient with antithrombin deficiency. Acta Cardiol 2008;63: McColl MD, Ramsay JE, Tait RC, Walker ID, McCall F, Conkie JA, et al. Risk factors for pregnancy associated venous thromboembolism. Thromb Haemost 1997;78: Sabadell J, Casellas M, Alijotas-Reig J, Arellano-Rodrigo E, Cabero L. Inherited antithrombin deficiency and pregnancy:

84 S78 Muszbek et al.: Antithrombin deficiency maternal and fetal outcomes. Eur J Obstet Gynecol Reprod Biol 2010;149: Tu CM, Hsueg CH, Chu KM, Cheng SM, Tsao TP. Simultaneous thromboses of double coronary arteries in a young male with antithrombin III deficiency. Am J Emerg Med 2009; 27:1169 e Peovska I, Maksimovic J, Kalpak O, Pejkov H, Bosevski M. Recurrent myocardial infarction in a young football player with antithrombin III deficiency. Cardiol J 2008;15: Roldan V, Ordonez A, Marin F, Zorio E, Soria JM, Minano A, et al. Antithrombin Cambridge II (A384S) supports a role for antithrombin deficiency in arterial thrombosis. Thromb Haemost 2009;101: Mahmoodi BK, Brouwer JL, Veeger NJ, van der Meer J. Hereditary deficiency of protein C or protein S confers increased risk of arterial thromboembolic events at a young age: results from a large family cohort study. Circulation 2008;118: Trenor CC 3rd, Michelson AD. Thrombophilia and pediatric stroke. Circulation 2010;121: Andrew M, Paes B, Milner R, Johnston M, Mitchell L, Tollefsen DM, et al. Development of the human coagulation system in the full-term infant. Blood 1987;70: Andrew M, Paes B, Milner R, Johnston M, Mitchell L, Tollefsen DM, et al. Development of the human coagulation system in the healthy premature infant. Blood 1988;72: Andrew M, Vegh P, Johnston M, Bowker J, Ofosu F, Mitchell L. Maturation of the hemostatic system during childhood. Blood 1992;80: Maclean PS, Tait RC. Hereditary and acquired antithrombin deficiency: epidemiology, pathogenesis and treatment options. Drugs 2007;67: Bushman JE, Palmieri D, Whinna HC, Church FC. Insight into the mechanism of asparaginase-induced depletion of antithrombin III in treatment of childhood acute lymphoblastic leukemia. Leuk Res 2000;24: Hernandez-Espinosa D, Minano A, Martinez C, Perez-Ceballos E, Heras I, Fuster JL, et al. L-asparaginase-induced antithrombin type I deficiency: implications for conformational diseases. Am J Pathol 2006;169: Hernandez-Espinosa D, Minano A, Ordonez A, Mota R, Martinez-Martinez I, Vicente V, Corral J. Dexamethasone induces a heat-stress response that ameliorates the conformational consequences on antithrombin of L-asparaginase treatment. J Thromb Haemost 2009;7: Abildgaard U, Lie M, Odegard OR. Antithrombin (heparin cofactor) assay with new chromogenic substrates (S-2238 and Chromozym TH). Thromb Res 1977;11: Odegard OR, Lie M, Abildgaard U. Antifactor Xa activity measured with amidolytic methods. Haemostasis 1976;5: Meijer P, Kluft C, Haverkate F, De Maat MP. The long-term within- and between-laboratory variability for assay of antithrombin, and proteins C and S: results derived from the external quality assessment program for thrombophilia screening of the ECAT Foundation. J Thromb Haemost 2003;1: Demers C, Henderson P, Blajchman MA, Wells MJ, Mitchell L, Johnston M, et al. An antithrombin III assay based on factor Xa inhibition provides a more reliable test to identify congenital antithrombin III deficiency than an assay based on thrombin inhibition. Thromb Haemost 1993;69: Conard J, Bara L, Horellou MH, Samama MM. Bovine or human thrombin in amidolytic at III assays. Influence of heparin cofactor II. Thromb Res 1986;41: Rossi E, Chiusolo P, Za T, Marietti S, Ciminello A, Leone G, et al. Report of a novel kindred with antithrombin heparinbinding site variant (47 Arg to His): demand for an automated progressive antithrombin assay to detect molecular variants with low thrombotic risk. Thromb Haemost 2007;98: Parvez Z, Fareed J, Messmore HL, Moncada R. Laser nephelometric quantitation of antithrombin-iii (AT-III) development of a new assay. Thromb Res 1981;24: Von Kaulla E, Von Kaulla KN. Antithrombin 3 and diseases. Am J Clin Pathol 1967;48: Marciniak E, Gockerman JP. Heparin-induced decrease in circulating antithrombin-iii. Lancet 1977;ii: Conard J. Inhibitors of blood coagulation. In: Thompson JM, editor. Blood Coagulation and Haemostasis, Vol. 2nd ed. Edinburgh: Churchill Livingstone, 1980: Laffan MaM R. Investigation of a thrombotic tendency. In: Lewis SM, Bain BJ, Bates I, editors. Practical Haematology, Vol. 10th ed. Philadelphia: Churchill Livingstone, 2006: Andersson G, Fagrell B, Holmgren K, Johnsson H, Ljungberg B, Wilhelmsson S. Antithrombin III in patients with acute deep vein thrombosis during heparin treatment (subcutaneous and intravenous) and during and after treatment with oral coumarins. Thromb Res 1984;34: O Brien JR, Etherington MD. Effect of heparin and warfarin on antithrombin III. Lancet 1977;2: Kottke-Marchant K, Duncan A. Antithrombin deficiency: issues in laboratory diagnosis. Arch Pathol Lab Med 2002; 126: Conard J. Antithrombin activity and antigen. In: Jespersen J, Bertina RM, Haverkate F, editors. Laboratory Techniques in Thrombosis A Manual 2nd Revised Edition of the ECAT Assay Procedures. Dordrecht: Kluwer, 1999:121 8.

85 Clin Chem Lab Med 2010;48(Suppl 1):S79 S by Walter de Gruyter Berlin New York. DOI /CCLM Review Factor V Leiden and FII testing in thromboembolic disorders Tadej Pajič* Department of Haematology, Division of Internal Medicine, University Medical Centre Ljubljana, Ljubljana, Slovenia Abstract Factor V Leiden and prothrombin (F2) c.20210g)a mutation detection are very important in order to define the increased relative risk for venous thromboembolism in selected patients. Use of DNA-based methods to detect both mutations has become widely available in clinical diagnostic laboratories, including fluorescence-based quantitative realtime PCR (qpcr). The latter is a rapid, simple, robust and reliable method to identify genotypes of interest. There are several chemistries used for qpcr; this article describes their principles and applicability for Factor V Leiden and prothrombin (F2) c.20210g)a mutation detection. Clin Chem Lab Med 2010;48:S Keywords: Factor V Leiden; FII 20210; quantitative realtime PCR; thromboembolic disorders. Introduction Venous thromboembolism (deep-vein thrombosis and pulmonary embolism, VT) is a major health problem in Western societies, with an incidence of about one per 1000 individuals, and increasing (1, 2). However, the risk for VT is agerelated, with the disease occurring primarily in older age groups (3). Although the major one, it is not the only risk factor related to VT. The pathogenesis of this complex disease involves circumstantial or acquired and genetic risk factors (4). Independent environmental factors contributing to VT include, but are not limited to: smoking, male gender, older age, immobilisation, surgery, malignant neoplasm, pregnancy, use of oral contraceptives, hormone-replacement therapy, and inflammatory conditions (3 6). Several genetic risk factors for venous thrombosis have been identified. Among them, antithrombin, protein C and protein S deficiency, dysfibrinogenemia and homozygous homocystinuria are rare, and account for only 5% 20% of patients with inherited thrombophilia. The situation changed remarkably in 1993 after the discovery of a defect in the *Corresponding author: Tadej Pajič, Zaloška 7, 1000 Ljubljana, Slovenia Phone/Fax: q , tadej.pajic@kclj.si Received September 10, 2010; accepted October 19, 2010 protein C anticoagulant pathway, which resulted in resistance to activated protein C (APC) (5, 7). Factor V Leiden (FVL) which causes resistance to APC was discovered in 1994 (8) and is the most common genetic risk factor for venous thrombosis. A few years later, the second most frequent and important inherited risk factor for thrombosis, prothrombin (F2) c.20210g)a mutation was identified (9). Mutation to FVL can be identified by DNA-based tests or by functional APC-resistance (APCR) tests. A DNA genetic assay is required for the identification of the prothrombin (F2) c.20210g)a mutation. Activated protein C resistance and Factor V Leiden mutation As described by Dahlback (6), the starting point for the discovery of APC resistance was unexpected behaviour in a functional assay for protein C observed in a plasma sample from a patient with thrombosis. The addition of APC to the patient s plasma did not result in prolongation of the clotting time. Further work demonstrated that APC resistance, as a phenotypic description of the condition, was inherited (9 11), and factor V was purified (12). Factor V is an important component of the coagulation cascade, which, in association with factor Xa, activates prothrombin to thrombin. In 1994, Bertina et al. (8) from the city of Leiden, The Netherlands, and other laboratories, reported (13 16) the same causative mutation in exon 10 of the gene encoding factor V(F5). This mutation resulted in a G A substitution at nucleotide 1691, producing a missense mutation that substitutes glutamine for arginine at amino-acid residue 506 (R506Q) in the factor V heavy chain protein product. Factor V has a 28-amino-acid leader peptide. Thus, this sequence variation has also been referred to as Arg534Gln wc.1601 G)A (p.arg534gln), FV Leiden mutation (FVL); rs6025x. The R506Q substitution in FVL involves one of the three sites on factor Va that is cleaved by APC; the other sites are at positions Arg306 and Arg679. The mutated protein is activated in a normal way and retains normal procoagulant activity, but is less susceptible to inactivation by APC and results in disposition to a hypercoagulable state (APC resistance) (17, 18). Moreover, the mutant factor V has diminished cofactor activity in the inactivation of factor VIIIa by APC (19, 20). Both these abnormalities result in the failure of APC to prolong the activated partial thromboplastin time, the classic APC resistance test. Depending upon the method used, the functional assay may pick up cases of APC resistance not due to FVL, although it accounts for more than 2010/528

86 S80 Pajič: FV Leiden and FII testing in thromboembolic disorders 90% of APC resistance (21 24). It is thought that the relatively common haplotype of the factor V gene, the HR 2 haplotype (H1299R), might also cause resistance to APC and increase the risk of venous thrombosis when co-inherited with FVL (25, 26). There is also a rare mutation in the second of the three sites in factor Va that APC cleaves which probably causes resistance to APC (27, 28). The prevalence of FVL mutation varies, with a higher frequency in Caucasians (up to 15%) and being rare or absent in Asians and Africans and in Australian and American natives (29 33). The FVL mutation was found to have high prevalence (in up to 50%) in patients with thrombosis, depending on the selection criteria used (10, 11, 33 35). The relative risk for venous thrombosis associated with the FVL mutation in the absence of other acquired or environmental predispositions is approximately four- to seven-fold for heterozygotes and 80-fold for homozygotes (25, 36). Prothrombin (F2) c.20210g)a mutation Prothrombin is an important component of the coagulation cascade. It is a vitamin K-dependent protein that participates in coagulation and its regulation. Prothrombin participates in the final stages of the blood coagulation cascade where it is converted to thrombin in the presence of factor Xa, factor Va, calcium ions, and phospholipids. A few years after identification of APC resistance and the causative FVL mutation, the second most frequent, and important inherited risk factor for thrombosis, was identified. The sequence alteration c.20210g)a is located in the 39- untranslated region of the prothrombin gene (F2) and is associated with slightly increased plasma prothrombin concentrations (37). The c.20210g)a mutation in F2 represents a gain-of-function mutation, causing increased recognition of the cleavage site, increased 39 end processing, and increased accumulation of mrna and protein synthesis (38). These changes can result in a hypercoagulable state. The relative risk for venous thrombosis associated with the mutation is two- to four-fold for heterozygotes (39, 40). Homozygosity for the c.20210g)a mutation is rare, but it increases the risk of thrombosis above that which has been observed for heterozygotes (5, 41). The prevalence of the mutation varies and is dependent on geographic location and ethnic origin. It is found in 0% 4% of the general population. The prevalence of the mutation in southern Europe is twice as high as in northern Europe. It is rare in Asian and African descendants, and in native Australians and Americans (33 35, 42). The mutation is also found in 6% 8% of patients with VTE (6, 37, 39, 43). It was found that 6% 12% of patients with VTE who were heterozygous for FVL also had the prothrombin mutation (F2) c.20210g)a (compound heterozygotes) (5, 44, 45). The relative risk for venous thrombosis in the individuals carrying both mutations are higher than in individuals without either mutation (25, 46, 47). Factor V Leiden and prothrombin (F2) c.20210g)a mutation detection The FVL mutation can be identified using a DNA-based test or by functional APC-resistance (APCR) tests. It is known that while errors can occur with genetic methods, testing for APC resistance can be helpful in assessing the presence of FVL, whether used initially as a screening test or if used in conjunction with molecular testing (48). A modification of the classic APC resistance test makes the test more sensitive to factor V and FVL mutations; with up to 100% sensitivity and specificity for the factor V mutation (49 53). A normal modified APCR test excludes the presence of FVL. However, when it is abnormal, the FVL genotype needs to be confirmed by genetic testing. A DNA assay is required for the identification of the prothrombin (F2) c.20210g)a mutation. As both mutations are very important in defining the increased relative risk for VTE in selected patients, the DNA-based methods that detect both mutations have become widely available in clinical diagnostic laboratories. Most assays are capable of using genomic DNA prepared from blood using a variety of extraction protocols. The various molecular methods in use for detecting the FVL and the prothrombin c.20210g)a mutations are also very sensitive, robust, accurate and reliable in identifying the mutation of interest (25, 48, 54). The gold standard method for detection of mutations is bidirectional sequencing of the specific genetic region of the gene of interest. However, there are many technologies that can be used to detect FVL and the prothrombin c.20210g)a mutation. The polymerase chain reaction (PCR), restriction-fragment-length polymorphism (PCR- RFLP) and allele-specific PCR are acceptable methods and used in many studies (48). However, in recent years the fluorescence-based quantitative real-time PCR (qpcr) approach (55 57) has become widely available in many clinical diagnostic laboratories. It has the advantage that it can detect and measure minute amounts of nucleic acids in a wide range of samples from numerous sources. In addition to its use as a research tool, many diagnostic applications have been developed (58), including FVL and prothrombin (F2) c.20210g)a mutation detection. This approach enables simple, fast, sensitive and specific nucleic acid quantification and mutation detection, particularly single base mutations (59 61). Some of the principle advantages and disadvantages of each will be discussed further. However, errors in DNA-based testing occur due to analytic mistakes as well as inadvertent sample mix-ups and transcription errors; an issue that has been discussed elsewhere (48, 62). Laboratory personnel need to be aware of the potential for these failures and for the rare silent mutations within the F5 gene (1692A)C, 1689G)A and 1696A)G) and rare sequence variations at or near the prothrombin (F2) c.20210g)a mutation (i.e., the mutations 20209C)T, 20207A)C, 20218A)G and 20221C)T). These will influence genetic analysis in a manner dependent upon the test system used (63). However, when these sequence variations

87 Pajič: FV Leiden and FII testing in thromboembolic disorders S81 are rare in the population their influence on the assay may be considered negligible (63). Real-time PCR There are several real-time PCR instruments and several different fluorescence detection technologies: DNA-binding agents, fluorescent primers and fluorescent probes. In the most common formats, the PCR reaction includes locus-specific primers in addition to a single or pair of fluorescencelabelled oligonucleotide probes. The specially designed primer systems amplify and detect the mutant and normal alleles using sequence-specific hybridisation-based assays. The most common probes that are used are dual hybridisation (LightCycler probes) or hydrolysis probes (58). One of the chemistries developed after the hydrolysis probes is Scorpions (64), and new chemistries are coming (59, 65). The experiment protocol usually combines purified genomic DNA, master mix (Taq DNA polymerase, buffer solution, deoxyribonucleotide triphosphates, and salts), primers and probe(s). This is followed by thermal cycling, reading, and analysis of the results. Mutation detection with melting curve analysis using dual hybridisation probes The DNA segment of interest is amplified with locus-specific primers, and the amplicon is detected by fluorescently labelled oligonucleotide probes (dual hybridisation probes) (58). Probes hybridise to an internal sequence of the amplified fragment during the annealing phase of the PCR cycle. One of the probes is labelled at the 59 end with LightCycler (LC) Red 640 or LC 705 (acceptor dye) (Roche Diagnostics, Mannheim, Germany). The second probe is labelled at the 39 end with fluorescein (donor dye). The 39 end of each probe is blocked with either a dye or a phosphate group to prevent extension during PCR. Only after hybridisation to the target sequence do the probes come in close proximity. In this situation, the energy emitted by the excitation of fluorescein is transferred to the acceptor dye, which then emits fluorescence at a longer wavelength, showing that fluorescence resonance energy transfer has occurred. The emitted fluorescence is then measured by the instrument (25, 66 68). Genotyping is performed using melting curve analysis with detection of the rate of melting of a wild-type probe from the amplicon after the amplification cycles when it is present at increased concentrations. Usually, the fluoresceinlabelled probe spans the mutation site (mutation probe). The stability of each probe/target complex is indicated by the melting temperature (Tm). Melting from the mutant allele occurs at a lower temperature than that from wild-type allele. The software plots the negative derivative of the fluorescence with respect to temperature. The peaks that are generated occur at Tms specific for the wild-type and mutant alleles (Figure 1). At a minimum, the analysis requires a heterozygous control and a no-template control (25, 63, 66 68). Interpretation of results is based on the presence or absence of peaks with a Tm that is specific for the wild-type or mutant allele. It is also very useful to check for the DTm wtm (wild-type) Tm (mutant)x which is less variable than the Tm values. In some circumstances, it helps to identify additional sequence variations (63) that need to be confirmed by other tests, frequently with sequencing. This approach in detecting FVL and the prothrombin (F2) c.20210g)a mutation is simple, allowing the result to be obtained in approximately 2 h and making analytical errors unlikely (63, 67, 68). The instrument software could have embedded algorithms to identify the genotype, making interpretation of the results easier. Although the instrument is relatively expensive and the method is prone to contamination with the amplicon or DNA (25, 63), it is a robust and reliable method for identifying the mutation of interest in the sample type to be used clinically. Mutation detection using hydrolysis probes Some systems use only a single fluorescently labelled oligonucleotide probe (TaqMan hydrolysis probes) that provide very sensitive and specific detection of DNA. The systems require a pair of PCR primers and a probe with both a reporter and a quencher dye attached (59, 69, 70). The quencher dye attached at the 39 end of the probe could be a fluorescent or non-fluorescent dye with a minor groove binder (MGB) at the 39 end (58, 59, 71). The latter increases the Tm of probes, allowing the use of shorter probes and, with the non-fluorescent quencher dye attached, provides more accurate allelic discrimination (72). The probe is designed to bind to the sequence amplified by the locus specific-primers (59). If the target sequence is present during the qpcr, the probe is hydrolysed with the 59-nuclease activity of the Taq DNA polymerase. Cleavage of the probe separates the reporter dye from the quencher dye, increasing the reporter dye signal and removing the probe from the target strand, allowing primer extension to continue to the end of the template strand. The fluorescent signal is generated and increased with each cycle (59, 71, 73). The number of PCR cycles necessary to detect a signal above the threshold is called the quantification cycle (Cq), and is directly proportional to the amount of target present at the beginning of the assay (58, 59, 71, 73). The change for signal corresponds to the increase in fluorescence intensity when the plateau phase is reached. Usually, the ROX (6- carboxy-x-rhodamine) fluorophore is added to the qpcr master mix as the passive reference for normalisation of the fluorescence. The use of ROX improves the results by compensating for small fluctuations in fluorescence due to things, such as bubbles and well-to-well variation that may occur in the plate (25, 71, 73). For genotyping, this technology uses two allele-specific hydrolysis probes and a PCR primer pair to detect the specific sequence variation. One probe is designed to detect the mutated allele and the other detect the wild-type. The allele-

88 S82 Pajič: FV Leiden and FII testing in thromboembolic disorders Figure 1 Factor V Leiden (A) and prothrombin (F2) c.20210g)a (B) mutation detection using melting curve analysis and dual hybridisation probes. The generated peaks occur at melting temperatures (Tms) specific for the mutant and wild-type alleles. (A) The first peak represents the mutant allele (578C"2.58C), the second relates to the wild-type allele (658C"2.58C), DTm wtm (wild-type) Tm (mutant)x is 88C"1.58C. (B) The first peak represents the mutant allele (498C"2.58C), the second one relates to the wild-type allele (598C"2.58C); DTm wtm (wild-type) Tm (mutant)x is 108C"1.58C. Red lines represent a heterozygous, light and dark green lines relate to a wild-type genotype. Blue line represents the no-template control. Factor V Leiden and prothrombin (F2) c.20210g)a mutation detection were performed using the Factor V Leiden Kit (66) and Factor II (Prothrombin) G201210A Kit, respectively (both Roche Diagnostics, Mannheim, Germany) (67). All reactions were performed in LightCycler capillaries using the LightCycler 2.0 instrument (Roche) with the LightCycler Software Version 4.1. The manufacturer (Roche) claims that genotyping with the LightCycler Software 4.1 will classify the rare silent mutations 1692A)C, 1689G)A and 1696A)G within the F5 gene as false-positive for Factor V Leiden (66). Within the F2 gene, in addition to the prothrombin (F2) c.20210g)a mutation, four known mutations at positions 20209, 20207, and exist (i.e., the mutations 20209C)T, 20207A)C, 20218A)G and 20221C)T), and these additional mutations are spanned by the mutation probe. These rare mutations will lead to an unknown or wild-type result after performing genotyping (67). specific hydrolysis probes have different reporter dyes. After PCR amplification, an endpoint plate reading is performed using a real-time PCR system. Allelic discrimination is based on the fluorescence measurements made during the plate reading to plot fluorescence values based on the signals from each well. The plotted fluorescence signals indicate which alleles are in each sample (25, 71). The other approach used to determine the specific allele is by using the quantification cycle (Cq). The principle of qpcr is the same as that described above, but genotyping of the sample relies on the Cq values of the specific probes for the mutated and normal alleles obtained during qpcr. The Cq values for the samples and controls are compared with the pre-defined Cq values of the specific probes for the mutated and normal alleles provided by the manufacturer (Figure 2) (74, 75). At a minimum, the analysis requires a heterozygous control and a no-template control. Using a 59-nuclease assay chemistry protocol is a simple way to obtain results in approximately 2 h. The advantage of the platform is that a large number of mutations can be detected simultaneously. The assays for FVL and F2 c.20210g)a mutation detection can be performed on the same microplate using the same thermal cycling conditions (71, 74, 75). The use of the constant and the qpcr appropriate amount of DNA extract from the samples and amplification controls in the amplification reaction is important, in order to prevent the problem of non-specific hybridisation or inhibition of fluorescence emission (71). Care must be taken with interpretation of results, as genotype are assigned manually and as additional sequence variation may suggest different genotypes (76) that need to be confirmed by another method. Mutation detection using Scorpion primers Scorpion primer technology allows precise discrimination between different alleles of a target nucleotide (64, 77 80). Scorpion primers combine a primer and a probe in a single molecule. There are two formats for Scorpions: uni- and bimolecular. The uni-molecular Scorpion format consists of a specific probe sequence that is held in a hairpin loop configuration by complementary stem sequences on the 59 and 39 sides of the probe. The fluorophore is attached to the 59 end and is quenched by a moiety coupled to the 39 end of the loop (64, 80). The hairpin loop is linked to the 59 end of

89 Pajič: FV Leiden and FII testing in thromboembolic disorders S83 Figure 2 Factor V Leiden (A) and prothrombin (F2) c.20210g)a (B) mutation detection using hydrolysis probes. In the assay, genotyping of the sample relies on the quantification cycle (Cq) values of the specific probes for the mutated (VIC probe) and normal alleles (FAM probe) obtained during qpcr (74, 75). The horizontal green line represents a threshold, set at a value of 0.2. It is specified by manufacturer Nanogen Advanced Diagnostics S.r.L., Torino, Italy. (A) The blue lines represent the amplification of the normal alleles (FAM probe). The red line represents amplification of the mutated alleles (VIC probe). The Figure show heterozygous and wild-type genotypes. The heterozygous genotype is identified when both probes (FAM and VIC) are amplified and the differences between Cq of the probes are -2 Cqs. The wild-type genotype is identified when the specific probe for the wild-type genotype is amplified. (B) The green lines represent amplification of the normal alleles (FAM probe). The blue line represents amplification of the mutated alleles (VIC probe). The Figure shows the heterozygous and wild-type genotype. The heterozygous genotype is identified when both probes (FAM and VIC) are amplified and the differences between Cq of the probes are -2 Cqs. The wild-type genotype is identified when the specific probe for the wild-type genotype is amplified. For the other pre-defined Cq values of the specific probes for the mutated and normal alleles, see references (74, 75). Delta Rn is difference in the intensity of the fluorescence signal of the interest and background. Factor V Leiden and prothrombin (F2) c.20210g)a mutation detection were performed using FACTOR V Q-PCR Alert Kit (74) and FACTOR II Q-PCR Alert Kit (75), respectively (both Nanogen Advanced Diagnostics S.r.L.). All reactions were performed in Micro-Amp optical 96-well plates using an ABI Prism 7000 Sequence Detection System with the Sequence Detection Software version (Applied Biosystems, Forster City, CA, USA). The manufacturer (Nanogen Advanced Diagnostics S.r.L.) does not specify the performance of the assay when rare mutations are present.

90 S84 Pajič: FV Leiden and FII testing in thromboembolic disorders a primer via a non-amplifiable monomer (blocker). After extension of the Scorpion primer, the specific probe sequence is able to bind to its complement within the extended amplicon, thus opening up the hairpin loop. The fluorescence is no longer quenched, which results in an increase in fluorescence from the reaction tube. The non-amplifiable monomer prevents non-specific PCR products (59, 64, 79, 80). Elements of the bi-molecular Scorpion are a PCR primer, blocker (a non-amplifiable monomer), and a probe with a fluorophore on one, and a dark quencher on a separate oligo. After extension of the Scorpion primer, the denaturation step is performed and the quencher oligo disassociates. When cooling is performed, the extended Scorpion with the specific probe sequence is able to bind to its complement and begins to fluorescence. The unextended primer is quenched (64, 79). The ability to multiplex greatly expands the power of qpcr analysis, particularly when applied to the simultaneous detection of sequence variations (58, 60, 61, 81). Simultaneous detection of both normal and mutant alleles in a single reaction is possible by combining two Scorpions in a multiplex reaction. Allelic discrimination can be achieved through hybridisation of the probe to the target sequence or by using the Amplification Refractory Mutation System (ARMS), where the primer is sited over the polymorphic site rather than the probe (79). This technology is used in a new assay kit for simultaneous FVL and F2 c.20210g)a mutation detection with the GeneExpert Dx System (Cepheid, Sunnyvale, CA, USA) using qpcr. The system automates and integrates sample purification, nucleic acid amplification, and detection of the target sequence in a single use disposable cartridge that contains the PCR reagents and hosts the PCR process. Each cartridge contains freeze-dried beads with all the necessary components for PCR: DNA polymerase, nucleotides, primers and Scorpions, that have to be rehydrated before use. The sample is added to the corresponding well in the cartridge and the analysis starts. Each Scorpion sequence is labelled with a specific fluorophore. Using the PCR cycles, the specific binding of the Scorpion sequence to the target mutation is detected by the system and the genotype is reported using algorithms imbedded in the software (Figure 3) which makes interpretation of the results straightforward (82). The newest instrument software version allows one mutation detection only, either FVL or F2 c.20210g)a. However, it is a simple and rapid assay because Scorpions have a fast reaction mechanism (59). The analysis can be performed using as little as 50 ml of fresh or frozen sodium citrate or EDTA anticoagulated whole blood. In our test, we obtained the result in approximately min. A disadvantage could be that the systems do not allow for high throughput sample processing. However, the analyses can be performed shortly after samples come to the laboratory; this could also be suitable for many coagulation laboratories that perform other thrombophilia testing. It is a one cartridge, one sample approach, and controls are not analysed at the same time. However, the assay includes a probe check control (Figure 3) which measures the fluorescence signal from the rehydrated probes, filling of Figure 3 Factor V Leiden and prothrombin (F2) c.20210g)a mutation detection using Scorpion primers and the GeneExpert Dx System (Cepheid, Sunnyvale, CA, USA). Factor V Leiden and prothrombin (F2) c.20210g)a mutation detection were performed using Xpert HemosIL FII&FV kit (Cepheid, Sunnyvale, CA, USA and Instrumentation Laboratory, Bedford, MA, USA). Following the PCR cycles, the specific binding of the Scorpion sequence to the target mutation is detected by the system and the genotype is reported using embedded algorithms wgene- Xpert Dx System version 2.1 (Cepheid)x (81). The manufacturer (Cepheid) states that rare factor V mutations (1692A)C, 1689G)A and 1696A)G) and any additional sequence variations in the probe binding region may interfere with detection of the target and produce an invalid result (81). the reaction tube, probe integrity and dye stability. The probe check passes the procedure if all criteria meet the assigned values (82). The external controls consisting of normal and abnormal whole blood samples may be used for quality control, including testing variability between different reagents. At the moment, I cannot find in the literature the performance of the assay when the rarely present additional sequence variations in the sample occur. Nevertheless, it is advisable to check any questionable results by other methods. Conclusions FVL and prothrombin (F2) c.20210g)a mutation detection are very important for defining the increased relative risk for VTE in selected patients. The DNA-based methods to detect both mutations have become widely available in clinical diagnostic laboratories. qpcr is a rapid, simple, robust and reliable method for identifying genotypes of interest. The newer instrument software with embedded algorithms to identify genotypes makes interpretation of the results easier, and possibly reduces the potential for error. However, care must be taken in interpreting questionable data which must be confirmed by others methods, preferably sequencing. Because the results of genetic analyses might have important clinical and family implications, it is important to perform