Thrombophilia Genetics

Joseph Emmerich

Martine Aiach

Pierre-Emmanuel Morange

Each year, approximately 1 in 1,000 individuals in industrialized countries develops deep-vein thrombosis (DVT) of the lower extremities.¹^,² Between 1% and 2% of these patients die from pulmonary embolism, and as many as 25% suffer the chronic effects of the postthrombotic syndrome. It is therefore important to identify high-risk patients with a genetic predisposition to thrombosis, particularly among those who have recurrent DVT. The terms “hypercoagulability” and “thrombophilia” refer to situations in which constitutional or acquired risk factors for thrombotic events are present.³

Hereditary thrombophilia is associated with a higher risk of thrombosis (mainly venous thrombosis) and is the antithesis of hemophilia, a disease associated with a high risk of spontaneous hemorrhage. Female carriers of hemophilia have clotting factor levels approximately 50% of normal and have a reduced risk of thrombosis.⁴ In contrast, thrombophilia is mainly an autosomal dominant trait linked to heterozygous loss-of-function mutations affecting inhibitors of coagulation or gain-of-function mutations affecting coagulation factors. Interestingly, half-normal coagulation factor levels are not associated with a bleeding tendency, whereas half-normal levels of coagulation inhibitors are associated with an increased risk of thrombosis.

As far back as 1856, Virchow postulated that thrombosis could be due to “changes in the composition of blood,” yet it was only in 1965 that Egeberg⁵ published the first case of inherited antithrombin (AT) deficiency, and this was followed by similar reports of protein C (PC) and protein S (PS) deficiency.⁶^,⁷^,⁸ Hereditary thrombophilia was initially considered as a rare monogenic disorder, but this view was challenged in the mid-1980s.⁹^,¹⁰ With the discovery of activated protein C resistance (APCR), a frequent factor V (FV) mutation known as FV_Leiden became a change of paradigm of genetic risk factors for thrombophilia and paved the way for studies of gene-gene and gene-environment interactions.¹¹^,¹² The main risk factors for venous thromboembolism (VTE) can be divided into acquired factors, genetic factors, and other factors that may vary due to genetic variations yet to be identified¹³^,¹⁴ (see Table 81.1). This chapter focuses on genetic risk factors, the main features of which are summarized in Table 81.2.

COAGULATION INHIBITOR DEFICIENCY

Antithrombin Deficiency

Biochemistry

AT is a 58-kDa plasma protein that regulates coagulation by inhibiting procoagulant serine proteases such as thrombin, activated (a) factor X, and factor IXa (see Chapter 18). AT belongs to the serpin family, members of which share structural features that allow them to form a stable stoichiometric complex with their target serine protease. The reaction between AT and thrombin involves the reactive center loop and the protease active site serine and is increased approximately 1,000-fold by heparin and other glycosaminoglycans (e.g., heparan sulfate) that are present on the endothelial surface. Heparin binding to AT involves amino-terminal amino acids 1 to 47, the A helix, and the D helix.¹⁵ AT is synthesized by the liver and circulates at a concentration of approximately 2.5 µmM; levels are decreased by estrogen and heparin therapy.

Clinical Manifestations

AT deficiency is transmitted as an autosomal dominant trait (0.02% prevalence in the general population) and is associated with a high risk of VTE (˜1% per year).¹⁶^,¹⁷^,¹⁸^,¹⁹ It is prevalent in <2% of unselected patients with a first VTE event, and it confers a higher risk of thrombosis than other thrombophilia. Homozygous AT deficiency may cause death in utero or severe, life-threatening thrombotic problems in the perinatal period.²⁰

Laboratory Diagnosis

There are two types of hereditary AT deficiency. Type I is the most frequent and is characterized by decreased activity in a heparin cofactor assay and a decreased protein concentration by immunoassay. Type II deficiencies are caused by functional defects, protein concentrations being normal or near normal in immunoassays. The dysfunction may affect the reactive site (IIRS) or the heparin-binding site (IIHBS) or both (IIPE) (pleiotropic effect).²¹ Homozygosity is mostly present in patients with type II_HBS’, with the exception of one patient with type II_PE deficiency.²⁰ Venous and arterial thrombosis may occur during infancy in such patients.²²^,²³

The ability of plasma to inhibit bovine thrombin and human FXa in the presence of heparin is the foundation for chromogenic heparin cofactor assays. Concentrations <80% (without heparin or estrogen treatment) call for further investigation, although only patients with severe deficiencies (<60%) are at high risk of thrombosis. Chromogenic assays distinguish type II deficiencies, in that type II_HBS has normal activity whereas type IIRS has low activity; the risk of VTE is very low in type IIHBS deficiency.²⁴

Molecular Basis

The human AT gene (SERPINC1—GenBank X68793.1) is located on chromosome arm 1q23-25 and it comprises seven exons spanning 1.3 kb and containing 10 Alu repeats.²⁵ A leader sequence of 32 amino acids is encoded by exon I and by the 5‘ end of exon II. The mature secreted polypeptide (432 amino acids) is encoded by exons II to VI. The reactive site, an Arg393-Ser394
bond, is located in the carboxy terminus of the protein and is encoded by exon VI. The HBS of AT is located in the amino-terminal domain and consists of two regions encoded by exon II and exon IIIa (see Chapter 18).

Table 81.1 Main risk factors for venous thrombosis

Acquired Risk Factors	Genetic Risk Factors	Other Risk Factors^a
Age	AT deficiency	Hyperhomocystinemia
History of venous thrombosis	Protein C deficiency	High levels of factor VIII
Surgery	Protein S deficiency	High levels of factor XI
Cancer	Factor V_Leiden	High levels of factor IX
Hormonal treatment	Factor II G20210A	High levels of factor VII
Antiphospholipid syndrome	Dysfibrinogenemia	High levels of TAFI
Myeloproliferative disorders	Fibrinogen γ10034T	Low levels of TFPI
Trauma	Non-0 blood group	APCR in the absence of FV_Leiden
Plaster cast
Obesity
TAFI, thrombin activatable fibrinolysis inhibitor; TFPI, tissue factor pathway inhibitor.
^aPossible genetic regulation.
From Chandler WL, Rodgers GM, Sprouse JT, et al. Elevated hemostatic factor levels as potential risk factors for thrombosis. Arch Pathol Lab Med 2002;126(11):1405-1414; Bertina RM. Genetic approach to thrombophilia. Thromb Haemost 2001;86(1):92-103, with permission.

One of the two alleles is not expressed in type I Deficiency, leading to a 50% reduction in the circulating protein. The database lists 80 different point mutations, including microinsertions and deletions; partial or entire deletions have been found in 12 patients.²¹) The high proportion of Alu repeats plays an important role in the generation of large SERPINC1 deletions causing AT type I Deficiency.

The type II_RS mutations result from amino acid substitutions in the reactive site loop, comprising amino acids 378 (P15) to 398 (P5‘), protruding from the surface of the protein. Mutations located in the hinge domain, affecting Ala382 or Ala384 (P10-P12), transform AT into a substrate for thrombin.²⁴ The Ala384 Ser mutation is particularly frequent in the British population.²⁵^,²⁶ Mutations affecting Gly392, Arg393, and Ser394 (P2, P1, P1′) prevent AT recognition by its target protease.²⁵

Most published mutations modifying AT affinity for heparin (type IIHBS)are located in the amino-terminal domain encoded by exon II or exon IIIa. Several patients have been found to
be homozygous for the Arg47Cys mutation or the Leu99Phe mutation.²¹^,²²

Table 81.2 Main features of hereditary thrombophilia

Protein Affected	Antithrombin	Protein C	Protein S	Factor V	Factor II
Gene location	1q23-25	2q13-14	3p11	1q21-22	11p11-q12
Type of mutation(s)		Loss of function Private mutations		Gain of function
Type of mutation(s)		Loss of function Private mutations		Arg506Gln	G20210A
Frequency in the general popult ation (%)	0.02^a	0.2-0.4^a	0.7-2.3^a	2-10	2-4
Type of assay	Heparin cofactor activity against FXa	Clotting assay or amidolytic assay	Clotting assay or immunoassay for free PS	APCR (second-generation aPTT-based assay) or FV genotyping	FII genotyping
Functional effect	Thrombin and FXa inhibitor	Reduce thrombin generation by inactivation of FVa and FVIIIa		FV variant resistant to APC inactivation	Increases the circulating FII concentration
Risk of VTE	× 10	× 4-5	× 4-5	× 4-5	× 3-4
PS, protein S; APCR, activated protein C resistance; aPTT, activated partial thromboplastin time; APC, activated protein C; VTE, venous thromboembolism. “From Seligsohn U, Lubetsky A. Genetic susceptibility to venous thrombosis. N Engl J Med 2001;344(16):1222-1231, with permission.

A cluster of mutations located on the carboxy-terminal side of the reactive loop, affecting residues 402, 404, 407, and 429, have a pleiotropic effect (type IIPE). The identification of an Asn187Asp mutation in a patient with recurrent thrombosis and apparently normal circulating AT is puzzling.²⁷ Indeed, this AT variant has a normal function in its native form, but tends to adopt an inactive conformation during storage in vitro, leading the authors to suspect that the thrombotic episodes were provoked by fever.

We identified a Phe229Leu mutation in a 13-month-old child with cerebral thrombosis. This mutation appears to generate an unstable variant that spontaneously polymerizes in the circulation.²³ This observation confirms that conformationally altered AT variants with reduced thermostability can be associated with severe thrombosis.²⁸

More than 130 different mutations have been reported and are described online at gene database (http://www1.imperial. ac.uk/departmentofmedicine/divisions/experimentalmedicine/haematology/coag/antithrombin/).

Protein C and Protein S Deficiencies

Biochemistry

PC is a vitamin K-dependent zymogen that is activated at the endothelial surface when thrombin binds to thrombomodulin. This reaction transforms thrombin from a procoagulant enzyme into an inhibitor, by activating PC to activated protein C (APC). PC binding to the endothelial protein C receptor (EPCR) augments its PC activation by the thrombin-thrombomodulin complex.²⁹ In the presence of its cofactor PS, APC degrades activated FV (FVa) and FVIIIa, thereby impeding further thrombin generation.³⁰ FV inactivation occurs in a biphasic reaction, with rapid cleavage at Arg506, followed by slower cleavage at Arg306. The first cleavage only partially affects FVa activity, whereas full inactivation occurs after the second cleavage at Arg306. PS markedly stimulates the second phase of the inactivation process, by a 20-fold enhancement of Arg306 hydrolysis.³¹ The mechanism of FVIIIa inhibition by APC is also biphasic, with cleavage at Arg562 and then at Arg336. FVIIIa inactivation by APC is increased by PS and FV, which act synergistically as cofactors for the reaction.³² The anticoagulant activity of PS has also been attributed to interaction with the prothrombinase complex, independent of APC³³^,³⁴ Furthermore, PS enhances around 10-fold the inhibition of factor Xa by tissue factor pathway inhibitor-alpha (TFPI-alpha) in the presence of Ca²⁺ and phospholipids.³⁵ In addition to its anticoagulant function, APC also binds to EPCR in lipid rafts/caveolar compartments to activate protease-activated receptor 1, thereby eliciting anti-inflammatory and cytoprotective signaling responses in endothelial cells. These properties have led to FDA approval of recombinant APC as a therapeutic drug for severe sepsis (see Chapter 123).

PC is synthesized by hepatocytes and circulates at a concentration of approximately 70 nM, with a half-life of approximately 8 hours. APC forms inactive complexes with serine protease inhibitors, mainly protein C inhibitor, but also protease nexin 3, α1 antitrypsin, and α2 macroglobulin.³⁶

Although PS is mainly produced by hepatocytes, it is also detected in endothelial cells and platelets. In the circulation, PS forms inactive complexes with C4b-binding protein (C4BP). Free PS represents approximately 40% of the total circulating level, and only this fraction has APC cofactor activity. C4BP is a multimeric protein composed of six or seven a chains, plus or minus a β chain. Only isoforms with a β chain (C4BP β+), which normally represent 80% of circulating C4BP, can bind PS. The interaction between PS and C4BP is reversible, but, in the presence of Ca²⁺, the dissociation constant is <10-⁹ mol/L. All circulating C4BP β+ molecules carry one molecule of PS, so free plasma PS results from a molar excess of PS over that of C4BP β+.³⁷ PS plasma levels are lower in women younger than 45 years and in those who are pregnant or are using oral contraceptives.³⁸ During acute-phase reactions, plasma C4BP concentrations increase after stimulation of the C4BP a, C4BP β, and PROS1 genes by inflammatory cytokines. PS not only suppresses coagulation as an enhancing cofactor for the coagulation inhibitors APC and TFPI but is also a physiologic ligand for the Tyro/axl/Mer-family of receptor tyrosine kinases that mediate an antiinflammatory regulatory loop of dendritic cell and monocyte inflammatory function.

Clinical Manifestations of Protein C and Protein S Deficiencies

PC and PS deficiencies are transmitted as autosomal dominant traits with incomplete penetrance, and heterozygous subjects belonging to families with the disorder are at risk of recurrent VTE during adulthood. Hereditary PC deficiency was first identified in subjects who had about half the normal PC concentration and a family history of thrombosis. At 45 years of age, 65% of affected members of the family described by Bovill et al.³⁹ and 50% of affected members of the 24 families described by Allaart et al.⁴⁰ were still free of thrombotic events. In prospectively studied asymptomatic members of thrombophilic families, the incidence of VTE was approximately 0.5% per patient-year in patients with PC deficiency and between 0.5% and 1.65% in patients with PS deficiency.¹⁴^,¹⁵^,¹⁶^,¹⁷^,¹⁸

The thrombotic risk associated with PC levels <67% was confirmed in a case-control study of unselected patients who developed DVT before 70 years of age, with a relative risk (RR) of approximately 3.⁴¹ Homozygous patients with undetectable PC have a very severe clinical phenotype, including life-threatening thrombotic complications at birth, mainly neonatal purpura fulminans with large bruises that become necrotic and gangrenous. The parents and family members of these homozygous infants have about half the normal PC concentration and are asymptomatic.⁴²^,⁴³ This form of genetically determined PC deficiency was believed to be recessively transmitted.

Heterozygous subjects belonging to families with PS deficiencies are at risk of recurrent thromboembolic disease in adulthood (review in⁴⁴). In heterozygous subjects, the probability of being free of thrombotic events at 45 years of age is approximately 50%.⁴⁵ We found that free PS and/or PS anticoagulant activity below the 10th percentile (ELISA < 75%) of control values was indeed associated with a risk of developing VTE.⁴⁶ Homozygous PS Deficiency, such as homozygous PC deficiency, is a rare disease associated with severe thrombosis, including neonatal skin necrosis and purpura fulminans.⁴⁷^,⁴⁸

Laboratory Diagnosis

Most clinical laboratories now use the snake venom protease Protac-based assay, allowing PC to be specifically and directly activated in plasma.⁴⁹ Such one-step assays evaluate the APC
generated after activation by Protac with synthetic substrates (amidolytic assays) or measure the prolongation of the activated partial thromboplastin time (aPTT) (anticoagulant assays). An immunoenzymatic assay measuring the protein concentration in plasma and functional assays measuring enzymatic or anticoagulant activity are used to distinguish several types of PC deficiency. In type I (quantitative) deficiency, which is caused by reduced synthesis of a normal protein, the plasma concentration is low in the three assays; this is the case in most PC deficiencies. Type II (qualitative) deficiency is characterized by normal synthesis of a nonfunctional protein that affects both the amidolytic and coagulation assays when the mutation disturbs the catalytic site but that affects only the coagulation assays when the mutation disturbs the interaction of PC with calcium, phospholipids, or macromolecular substrates (FV and FVIII). Therefore, it is recommended to use coagulation assays to screen patients for PC deficiency. It is difficult to establish normal ranges, as PC levels in subjects with PC gene abnormalities overlap with levels in healthy subjects⁴⁹ and vary with age. According to Miletich,⁴⁹ the increase in the PC concentration is approximately 4% per decade. Therefore, patients with PC concentrations <70% may have a hereditary deficiency, although values between 55% and 70% must be considered as borderline.

The diagnosis of PS deficiency is complicated by the presence in plasma of two molecular forms, that is, free PS and C4BP/PS complexes. Therefore, to measure the total circulating PS, immunoenzymatic assays have to be performed in conditions in which C4BP/PS complexes are dissociated, such as with highly diluted plasma in anti-PS-coated plates.⁵⁰ PS deficiency characterized by a low free PS concentration but a normal total PS concentration was identified by Comp.⁵⁰ A monoclonal antibody-based immunoenzymatic assay is available to measure free PS specifically.⁵¹ APC cofactor activity can be evaluated in an aPTT assay after adding diluted plasma to PS-depleted plasma in the presence of purified APC and purified FVa.

According to the International Society on Thrombosis and Hemostasis (ISTH) standardization subcommittee, three types of PS deficiency have been defined on the basis of total PS levels, free PS levels, and APC cofactor activity. Type I deficiency is characterized by low total PS and free PS antigen levels; type II deficiency is characterized by normal free PS and low APC cofactor activity; and type III PS deficiency is characterized by low free PS levels and normal or near-normal total PS levels. Type I and type III deficiencies in fact appear to be two phenotypic expressions of the same genetic disease,⁴⁵ and free PS is used to diagnose PS deficiency in these cases. The lower normal limit of total and free PS levels is 65% of the level observed in a pool of normal plasmas. However, the reference range in women younger than 45 years is approximately 55% under the same conditions. It is recommended to use both the clotting assay and the monoclonal based immunoassay specific for free PS to screen patients for PS deficiency.

Molecular Bases of Protein C andProtein S Deficiencies

The human PC gene (PROC) maps to chromosome arm 2q13-q14, spans over 11 kb, and comprises a coding region (exons II to IX) and a 5‘ untranslated region encompassing exon I.⁵² The protein domains encoded by exons II to IX show considerable homology with other vitamin K-dependent coagulation proteins such as factors VII, IX, and ×X. Exon II codes for a signal peptide, whereas exon III codes for a propeptide and a 38-amino acid sequence containing nine Glu residues. Exons IV, V, and VI encode a short connecting sequence and two epidermal growth factor (EGF)-like domains, respectively. Exon VII encodes both a domain encompassing a 12-amino acid activation peptide that is released after activation of PC by thrombin and dipeptide 156 to 157, which, when cleaved, yields the mature two-chain form of the protein. Exons VIII and IX encode the serine protease domain, with His211, Asp257, and Ser360 forming the catalytic triad.

The database published on behalf of the ISTH coagulation inhibitor subcommittee⁵³ lists 160 different mutations. The proportion of missense mutations is very high and the spectrum very wide, making the molecular basis of PC deficiencies similar to that of FIX deficiency (hemophilia B). The large spectrum of mutations responsible for PC deficiency is probably caused by a high rate of de novo mutations.⁵⁴ Among 90 patients with point mutations, selected on the basis of PC levels <65% and with at least one episode of thrombosis, 76 bore a missense mutation. Interestingly, 4.4% of the patients with the missense mutation were homozygous or compound heterozygotes, of whom only one had purpura fulminans at birth.⁵⁵ Homozygosity and compound heterozygosity may account for concentrations ranging from <1% to 25%. Only patients with concentrations <5% are at risk of purpura fulminans.⁵⁵^,⁵⁶

The amount of PC produced by the mutant allele (null or plus), as well as genetic status (heterozygous, homozygous, or compound heterozygous), partly accounts for the variable clinical expression. However, these factors do not explain why, in many families carrying a single PC gene mutation, more than 50% of the affected members remain asymptomatic. Other putative genetic factors may therefore favor thrombosis or protect against disease expression.

Two kinds of type II deficiency can be distinguished on the basis of plasma assays after PC activation by Protac. The substitution of different residues in the propeptide or the gamma carboxyglutamic acid (GLA) domain always resulted in low PC anticoagulant activity, whereas amidolytic activity was normal. It is not surprising that mutations giving rise to this biologic phenotype affect exon III because both the propeptide and the N-terminal region play a major role in the formation of the GLA domain, required for the anticoagulant activity of PC.

A few mutations in exon IX, which encodes the serine protease domain, also affected the coagulation assay but not the amidolytic assay, suggesting that, in addition to the catalytic pocket, this domain encompasses a region or regions required for PC anticoagulant activity. The other mutations in exon IX led to abnormal results in both the coagulation assay and the amidolytic assay⁵⁷^,⁵⁸ The possible structural impact of natural substitutions has been examined by Greengard et al.⁵⁹ using three-dimensional molecular modeling.

Two homologous genes for PS map to chromosome 3p11.⁵² The active gene, PROS1, spans over 80 kb and comprises 15 exons. Because PROS2 has no open reading frame and shows multiple base changes, stop codons, and frameshifts, it is probably a pseudogene. The 5‘ part of the PROS1 gene shows strong homology with the other vitamin K-dependent proteins, particularly PC. Exon I encodes the signal peptide; exon II encodes the propeptide and the GLA domain; exon III encodes a domain with a high aromatic amino acid content; exon IV encodes a thrombin-sensitive loop; and exons V to VIII encode four EGF-like domains. The 3‘ part of the PROS1 gene
differs from that of all other known coagulation proteins: The last seven exons (IX to XV) encode a sex-hormone-bindingglobulin-homologous domain.⁵⁶ On screening consecutive patients with unexplained thrombosis and low PS levels, we found mutations in 70% of cases.⁶⁰ Simmonds et al.⁶¹ found a mutation in 41% of 34 patients with type I deficiency, selected using criteria similar to ours. The mutations observed in type I deficiency are distributed throughout the coding sequence. More than 200 different mutations are listed in the database.⁴⁴^,⁴⁸^,⁶² Other mechanisms may explain why one allele is not expressed in patients with type I deficiency who have no detectable mutation. Because the screening strategy is based on selective amplification of the PROS1 gene, recombination events between the PROS2 and PROS1 genes are not detected by polymerase chain reaction (PCR)-based techniques. However, only three PS gene abnormalities involving large deletions have been shown to be responsible for PS deficiencies.⁶³ No mutations in the promoter domain have been identified (unpublished data).

A single mutation transforming Ser460 to Pro (Heerlen polymorphism), first described by Bertina et al.⁶⁴ as a polymorphism, was found in 28 of our 118 patients referred with unexplained thrombosis and PS type III deficiency; however, approximately 50% of the patients also carried the FV Arg506Gln mutation or a PC gene defect, suggesting a cooperative effect on clinical expression.⁶⁰ The fact that the affected members of seven families carrying the Ser460Pro mutation were all asymptomatic⁶⁵ further suggests that this mutation is not itself a major cause of thrombosis. The type I and type III phenotypes were found to be associated with a similar thrombotic risk, and the two phenotypes coexisted in 14 families, leading the investigators to postulate that type I and type III are phenotypic variants of the same genotype.⁴⁵ Taken together, these results show that type III phenotypes have a heterogeneous molecular basis and a wide range of clinical consequences. Recently, it has been shown that the cutoff levels of free PS identifying type III individuals at risk for DVT might be far below the normal range in healthy volunteers. Indeed, individuals with free PS level less than the 5th percentile (<41 IU/dL) had 5.6 times increased risk of DVT compared with those in the upper quartile (>91 IU/dL).⁶⁶

PS type II deficiency is fairly infrequent, and only a few mutations have been identified in patients with normal PS concentrations and low APC cofactor activity. Most mutations giving rise to a type II phenotype are located in the amino-terminal part of PS, which is homologous to that of other vitamin K-dependent proteins and encodes the domains interacting with APC.⁶⁰

A database of mutations responsible for defective genes, including PROC and PROS1, is available online at http://www. hgmd.cf.ac.uk.

FACTOR V_LEIDEN

Biochemistry of Activated Protein C Resistance

In 1993, Dahlback described three families in whom APC did not yield the expected prolongation of the clotting time in an aPTT assay⁶⁷^,⁶⁸; this defined a new phenotype called APCR. This was found in 15% to 25% of patients with DVT and in 2% to 10% of control subjects.⁶⁹^,⁷⁰ The study by Bertina et al.⁷¹ showed that APCR cosegregated with the FV gene and with a single mutation (FV_Leiden, Arg506 Gln) affecting one of the APC cleavage sites. Most, but not all, cases of APCR are caused by FV_Leiden. In the Leiden Thrombophilia Study (LETS), after exclusion of patients with FV_Leiden, a relation was observed between APC sensitivity and the risk of thrombosis: the lower the normalized APC sensitivity ratio, the higher the associated risk. The adjusted (i.e., age, sex, and FVIII) odds ratio (OR) for the lowest quartile was 2.5 (95% confidence intervals [CI], 1.5 to 4.2) compared with the highest quartile.⁷² In another study, phenotypic resistance to APC was associated with VTE, independently of FV_Leiden status; the age- and sex-adjusted OR for VTE was 1.7 (95% CI, 1.0 to 2.7) in participants who had a normalized APC sensitivity ratio of 0.50 to 0.84 compared with those who had a ratio of 0.85 to 1.3.⁷³ Clinical states with low APC sensitivity that are not caused by FV_Leiden may also be acquired, as is the case during pregnancy, with the use of oral contraceptives, or in patients with lupus anticoagulant or high levels of FVIII.

Molecular Basis of Activated Protein C Resistance

FV is a 330-kDa multidomain single-chain glycoprotein, with a plasma concentration of 20 nmol/L (0.007 g/L).⁷⁴^,⁷⁵ The FV gene (gene locus on chromosome arm 1q23) spans more than 80 kb and contains 25 exons. The complimentary DNA (cDNA) has a length of 6,672 bp and encodes a preprotein of 224 amino acids. Similar to FVIII, FV is organized into six domains (i.e., A1, A2, B, A3, C1, and C2). FV and FVIII share approximately 40% sequence identity in their A- and C-domains. Thrombin and FXa activate FV by a cleavage at peptide bonds at positions 709, 1,018, and 1,545, thereby releasing the B-domain, which connects the heavy chain (domains A1-A2) to the light chain (domains A3-C1-C2). Upon activation, FVa is formed by the heavy and light chains that are noncovalently associated by a Ca²⁺ ion. FVa is an essential FXa cofactor; its presence in the prothrombinase complex enhances the rate of prothrombin activation 10³– to 10⁵-fold. Downregulation of the procoagulant activity of FVa is accomplished through its inactivation by APC at positions Arg306, Arg506, and Arg679. Cleavage at Arg506 is essential for optimal exposure of cleavage sites Arg306 and Arg679 but results in partial inactivation of FVa (˜40% of procoagulant activity remains). The slower Arg306 cleavage is required for complete inactivation of the protein, whereas the third cleavage site (Arg679) is less important. Therefore, any defect on one or more of these three cleavage sites (i.e., Arg506, Arg306, and Arg679) may potentially affect APC inactivation. FVa inactivation is enhanced by PS. The PS effect is highly phospholipid dependent and is mediated by a selective 20-fold stimulation of APC-catalyzed cleavage at Arg306. In addition to its major role in the procoagulant process when activated, FV also has an anticoagulant role as an APC cofactor (such as PS), downregulating FVIIIa activity after cleavage at position 506. Therefore, this dual pathway involving both inactivation of procoagulant FVa by APC and partial proteolysis of intact FV generating an anticoagulant protein is a complex and subtle regulatory system.

Bertina identified the genetic explanation for the APCR phenotype first in 1994, followed closely by three other groups.⁷¹^,⁷⁶^,⁷⁷^,⁷⁸ A single base mutation, guanine to adenine at position 1,691 of the FV gene, is responsible for the Arg506Gln mutation known as FV_Leiden. This mutation results in a substantially reduced
anticoagulant response to APC, because FV_Leiden is inactivated about 10 times slower than normal FV. This impairment of FVa inactivation increases thrombin generation and explains more than 90% of clinical APCR phenotypes. However, other mechanisms may contribute to the hypercoagulable state, as suggested by similar inactivation of FVa and FVa_Leiden by APC in the presence of PS, FXa, and high concentrations of FVa.⁶⁸ The fact that FV_Leiden is a much less active cofactor of APC than wild-type FV for FVIIIa inactivation may also explain the thrombophilic state. Impaired FVa inactivation and loss of APC cofactor activity contribute equally to the APCR phenotype in subjects with FV_Leiden.⁷⁹ Interestingly, two other mutations in FV affect the Arg306 cleavage site. Arg306 is replaced by Gly in FV_Hong κKong, and by Thr in FV_Cambridge.⁸⁰^,⁸¹ FV_{Hong Kong} is prevalent (5%) among the Chinese in Hong Kong, but neither of these latter mutations is associated with an increased risk of developing venous thrombosis.⁸² The two FV mutations involving Arg306 yield identical mild APCR patterns. The associated APC resistance is due not only to incomplete FVa inactivation, but also to reduced APC cofactor activity in FVIIIa inactivation.⁷⁵ It strongly argues for the importance of the loss of the APC cofactor activity of FV when it is mutated; it is also strengthened by the demonstration that a recombinant FVa, in which the three known APC cleavage sites have been mutated, is still inactivated by APC, although much slower than the wild-type protein.⁸³ FV Ile359Thr (FVLiverpool) and FV Glu666Asp are two rare mutations also associated with thrombosis and APCR.⁷⁵

Epidemiology

The prevalence of FV_Leiden is high in populations of white descent, but low in native populations of Asia, Africa, and Australia.⁸⁴ All FV_Leiden alleles are carried by the same haplotype, leading to the inference that the mutation occurred only once and spread by a founder effect. The estimated time of the mutation is approximately 30,000 years ago, implying that it took place after the out-of-Africa divergence that occurred approximately 100,000 years ago.⁸⁵ Its spread among whites and its high prevalence suggest that FV_Leiden is associated with a survival advantage, such as a decrease in severe bleeding after delivery⁸⁶^,⁸⁷

The frequency of FV_Leiden in white populations is between 2% and 15%.⁶⁹^,⁷⁰^,⁸⁸^,⁸⁹^,⁹⁰^,⁹¹^,⁹²^,⁹³^,⁹⁴^,⁹⁵ FV_Leiden is very rare in China, Korea, Taiwan, and Japan.⁹⁶^,⁹⁷^,⁹⁸ In the United States, FV_Leiden is found in 2.21% of Hispanic Americans, in 1.23% of African Americans, in 0.45% of Asian Americans, and in 1.25% of American Indians.⁹⁹ The risk of DVT in heterozygous carriers is approximately fivefold higher than in control population (see Table 81.3). The association of FV_Leiden with pulmonary embolism is much weaker than with DVT, with a prevalence of FV_Leiden <10% versus 15% to 25%,¹⁰⁰ perhaps explained by the formation of a thrombus that is more stable and adherent and less prone to embolism.¹⁰¹^,¹⁰² Whatever be the reasons, this imbalance between the risk of DVT and pulmonary embolism, known as FV_Leiden paradox, is not caused by a bias in selection and is specific for FV_Leiden, as it has not been observed in carriers of the prothrombin G20210A mutation.¹⁰² To date, there is no clear explanation to understand the FV_Leiden paradox.¹⁰³

Only gold members can continue reading. Log In or Register to continue