Von Willebrand Disease, Type 1

A number sign (#) is used with this entry because von Willebrand disease (VWD) type 1 is caused by heterozygous mutation in the gene encoding von Willebrand factor (VWF; 613160), which maps to chromosome 12p13. VWD type 2 (VWD2; 613554) and VWD type 3 (VWD3; 277480) are also caused by mutation in the VWF gene.

Description

Von Willebrand disease is the most common inherited bleeding disorder. It is characterized clinically by mucocutaneous bleeding, such as epistaxis and menorrhagia, and prolonged bleeding after surgery or trauma. The disorder results from a defect in platelet aggregation due to defects in the von Willebrand factor protein. Von Willebrand factor is a large, multimeric protein that plays a role in platelet adhesion and also serves as a carrier for the thrombotic protein factor VIII (F8; 300841). F8 is mutated in hemophilia A (summary by Goodeve, 2010).

For a review of the various forms of von Willebrand disease, see Leebeek and Eikenboom (2016).

Classification of von Willebrand Disease

The classification of von Willebrand disease has a long and complex history. The current classification is based on that described by Sadler (1994) and updated by Sadler et al. (2006), which delineates 3 main subtypes according to the mutant protein phenotype. An earlier classification developed by a working party of the European Thrombosis Research Organization was provided by Zimmerman and Ruggeri (1983).

Von Willebrand Disease Type 1

VWD type 1 is a quantitative partial deficiency of circulating VWF. In this type of VWD, there is a normal ratio of functional VWF activity (VWF:RCo, ristocetin cofactor activity) relative to VWF antigen level (VWF:Ag) (Sadler et al., 2006, Goodeve, 2010). Mannucci (2004) stated that type 1 VWD accounts for 60 to 80% of all VWD cases and is characterized by mild to moderate quantitative deficiencies of VWF and factor VIII, which are coordinately reduced to 5 to 30% of normal plasma levels (pathogenic levels of 5 to 30 IU/dL). In an updated consensus statement, Sadler et al. (2006) noted that (1) some cases of VWF type 1 may have subtle abnormal VWF multimer patterns, but still retain normal functional activity, and (2) that loci other than VWF may be responsible for some cases of VWD.

In reviews, James and Lillicrap (2008) and Lillicrap (2009) stated that the knowledge of the pathogenesis and molecular basis of type 1 VWD is still in its infancy and still evolving. Population studies have indicated that type 1 VWD is a complex genetic trait associated with a variety of genetic and environmental factors, and that additional loci in addition to VWF are likely involved. There is still uncertainty about the pathogenicity of many identified putative VWF variants, and the incomplete penetrance and variable expressivity of type 1 disease contributes to complexity in diagnosis and understanding of disease pathogenesis.

Von Willebrand Disease Type 2

VWD type 2 (613554), which accounts for 10 to 30% of cases, is characterized by qualitative abnormalities of VWF; it is further divided into subtypes 2A, 2B, 2M, and 2N. The mutant VWF protein in types 2A, 2B, and 2M are defective in their platelet-dependent function, whereas the mutant protein in type 2N is defective in its ability to bind F8 (Mannucci, 2004; Sadler et al., 2006; Goodeve, 2010).

Von Willebrand Disease Type 3

VWD type 3 (277480), which accounts for 1 to 5% of cases, is characterized by a severe quantitative defect of VWF in plasma (less than 1% of normal plasma levels), with low but usually detectable levels of factor VIII (1 to 10% of normal plasma levels). In the rare type 3 disease (1 in 1 million people), symptoms are more frequent and severe (Mannucci, 2004, Sadler et al., 2006).

Clinical Features

Type 1 VWD is the most frequent type of von Willebrand disease. However, laboratory aspects of diagnosis rely on phenotypic assays of VWF which have an uncertain relationship with VWF function in vivo and with clinical bleeding. Type 1 VWD is characterized by mild to moderate reductions of plasma VWF levels, which can vary with time, and plasma levels of VWF vary widely in the normal population, ranging from less than 50 U/dl to about 200 U/dl. The molecular diagnosis of type 1 is poorly understood. All of these factors make the definitive diagnosis of type 1 VWD particularly difficult (O'Brien et al., 2003; Cumming et al., 2006).

O'Brien et al. (2003) reported 12 probands, 10 Canadian and 2 from the U.K., with a clinical diagnosis of type 1 VWD. All had a bleeding history with epistaxis, menorrhagia, easy bruising, and excessive bleeding from wounds, especially after surgery and dental procedures. Laboratory studies showed that all had VWF multimers of various sizes, but at reduced intensity, consistent with a quantitative defect in type 1 VWD. Testing for VWF:Ag (VWF protein antigen levels) and VWF:RCo (VWF platelet-induced aggregation function with ristocetin) showed decreased levels of both compared to normal, and FVIII:C (factor 8 pro-coagulant activity) was normal in all except 1 patient. Notably, all except 1 had type O of the ABO blood group. All patients were found to have the same heterozygous mutation in the VWF gene (Y1584C; 613160.0029).

Cumming et al. (2006) examined 40 U.K. families with a clinical diagnosis of type 1 VWD. After review, 6 were found to have type 2 VWD and 2 additional families did not have VWD. Genetic analysis identified 6 pathogenic VWF mutations in 9 (28%) of the 32 remaining families with type 1 VWD. Three index cases (9%) had more than 1 candidate mutation. Fifteen (47%) families had no identified mutation. Cumming et al. (2006) identified the Y1584C change (613160.0029) in 8 (25%) families, but the variant segregated with the disease in only 3 families, did not segregate in 3 families, and yielded equivocal results in 2 families, suggesting that it is a common polymorphism that increases susceptibility to the disease, but does not cause the disease by itself. The authors concluded that mutation screening of the VWF gene has limited general utility in genetic diagnostic and family studies in type 1 VWD due to incomplete penetrance and variable expressivity.

Goodeve et al. (2007) identified VWF mutations in 105 (70%) of 150 index patients from 9 European countries with a clinical diagnosis of type 1 VWD. A total of 123 candidate VWF mutations were identified, including 53 novel mutations: 88 patients had 1 mutation, 16 had 2, and 1 had 3. However, multimer analysis showed that a substantial part of the cohort (38%, 57 of 150) had an abnormal multimer pattern that did not fit into the accepted criteria for type 1 VWD. In addition, 18 patients had decreased binding of F8 to VWF, including 3 with markedly reduced binding consistent with type 2N VWD. Fifty-one probands with normal multimers had a VWF mutation, and 42 with normal multimers had no identifiable VWF mutations, indicating genetic heterogeneity. Incomplete segregation was observed in many of the families with mutations.

James et al. (2007) examined 123 Canadian families with a clinical diagnosis of type 1 VWD. All patients had excessive mucocutaneous bleeding, including menorrhagia, easy bruising, epistaxis, postdental procedure bleeding, and postoperative bleeding. The mean VWF:Ag level for the index cases was 0.36 IU/mL, the mean VWF:RCo level was 0.34 IU/mL, and the mean FVIII level was 0.54 IU/mL. The VWF multimer pattern was normal. Most of the index cases (77, or 62%) were blood group O (see 616093). Putative VWF mutations were found in 78 patients (63%), leaving 37% with no identified changes. The changes comprised 50 different variants: 31 (62%) missense mutations, 8 (16%) changes involving the VWF transcriptional regulatory region, 5 (10%) small deletions/insertions, 5 (10%) splice site mutations, and 1 nonsense mutation. Twenty-one of the index cases had more than 1 putative VWF mutation, although many of these genetic changes may have been polymorphisms. James et al. (2007) noted that sequence variations within the VWF gene were clearly not the sole genetic determinants of the type 1 VWD phenotype in some cases. In addition, the relative contribution of VWF sequence variation appeared to be most important in more severely affected individuals, and the contribution of factors such as ABO blood group became more significant in milder cases. The findings indicated that more severe cases of type 1 VWD tend to have mutations within the VWF gene that are highly heritable. In contrast, milder cases, where heritability is variable, have more complex genetic determinants, and are more likely to have contributions from other genetic factors, such as ABO blood group, as well as possibly environmental factors.

Relationship to Blood Group O

Gill et al. (1987) found a significant association between ABO blood group O (see 616093) and decreased levels of VWF antigen (VWF:Ag) among 1,117 volunteer blood donors analyzed by quantitative immunoelectrophoresis. Individuals with blood group O had the lowest mean VWF:Ag levels (74.8 U/dL), followed by group A (105.9 U/dL), group B (116.9 U/dL), and group AB (123.3 U/dL). In addition, multiple regression analysis revealed that age significantly correlated with VWF:Ag levels in each blood group. Among 114 patients with type 1 VWD, blood group O was found in 88 (77%), group A in 21 (18%), group B in 5 (4%), and group AB in none (0%), whereas the frequency of these blood groups in the normal population was significantly different (45%, 45%, 7% and 3%, respectively). In contrast, patients with VWD type 2 or type 3 had ABO blood group frequencies that were not different from the expected distribution. Gill et al. (1987) concluded that there may be a subset of symptomatic type 1 VWD patients with decreased concentrations of structurally normal VWF due to the presence of blood group O. Similarly, some individuals with blood group AB and a genetic defect in VWF may not be diagnosed as affected because VWF levels are elevated due to their blood type.

Plasma VWF levels may be modified by genetic and environmental factors such as thyroid hormones, estrogens, or stress. The best-characterized genetic modifier is the ABO blood group, which accounts for approximately 30% of the genetic effect (Nichols and Ginsburg, 1997). See also Nitu-Whalley et al. (2000) concerning the influence of the ABO blood group in type 1 VWD. In the study of Casana et al. (2001), the number of patients with blood group O was regarded as significant in the milder deficiency group with variable expression of the gene, i.e., blood group may be irrelevant in families with complete penetrance, but might be a factor in cases with mild phenotypes and incomplete penetrance. Mutations affecting genetic modifiers might also cause type 1 VWD, even in the absence of a mutation at the VWF gene, which might be the case for the families in which no linkage was obtained by Casana et al. (2001).

The effect of blood group O on VWD appears to be through reduced survival, or increased clearance, of VWF, which may be related to different glycosyltransferase alleles that determine the ABO blood group. ABO antigens are added to the N-linked oligosaccharide chains present in mature VWF, and this glycosylation may protect VWF from clearance. However, the glycosyltransferase is nonfunctional in blood group O due to a null allele. Individuals with blood group O would thus have a lack of glycosylation of VWF, which would result in a loss of protective carbohydrate structure and increased clearance of VWF (review by Goodeve, 2010).

Acquired Von Willebrand Disease

A phenocopy of von Willebrand disease in a patient with the autoimmune disorder systemic lupus erythematosus (SLE; 152700) was reported by Simone et al. (1968). Acquired von Willebrand disease may have an autoimmune basis, the target of immunologic damage being endothelial cells that synthesize von Willebrand factor (Wautier et al., 1976).

Other Features

Early reports described disorders associated with von Willebrand disease. Quick (1967) reported a mother and daughter with telangiectasia and von Willebrand disease. Association with angiodysplasia of intestinal vessels, in some cases demonstrable by arteriography and visible at surgery, has been described (Ramsay et al., 1976).

Kernoff et al. (1981) found extensive atherosclerosis in an elderly man with von Willebrand disease, suggesting that the defect in platelet adhesion does not interfere with the atherosclerotic process.

Pathogenesis

Several observations (Cornu et al., 1963; Biggs and Matthews, 1963) are pertinent to the nature of the factor VIII defect in von Willebrand disease: (1) Blood from a patient with hemophilia A (306700), due to a defect in the F8 gene, will correct the clotting defect in von Willebrand disease. (2) The converse is not true: blood from a patient with von Willebrand disease will not correct the clotting defect in hemophilia A. (3) The bleeding tendency in von Willebrand disease is corrected promptly by normal blood. (4) After administration of hemophilia A blood to von Willebrand patients there is a delay of several hours before the level of F8 reaches normal. These observations indicated a relationship between VWF and F8.

In 3 patients with von Willebrand disease, Gralnick and Coller (1976) found that the factor VIII-von Willebrand factor protein was present in normal amounts and had normal procoagulant and antigen activities. The protein was, however, deficient in both carbohydrate and von Willebrand factor activity. The carbohydrate portion of this glycoprotein seemed to be essential to its interaction with platelets or blood vessel wall or both.

Nachman et al. (1980) studied the factor VIII antigen molecule in classic type 1 and variant type 2A. The 2-dimensional peptide maps were 'remarkably similar' both to each other and to normal factor VIII. Thus, the differences observed in von Willebrand disease were probably not due to qualitatively abnormal molecules but rather to 'quantitative shifts in the metabolism of normal factor VIII antigen molecules.' Type 1 is usually associated with a concordant decrease in VIII:C, VIII:A, and VIII:VWF, whereas type 2A shows a disproportionate decrease in VIII:VWF.

Clinical Management

Patients with type 1 VWD have a high probability of responding to the vasopressin analog desmopressin acetate (1-desamino-8-D-arginine vasopressin; dDAVP), which raises the level of factor VIII/von Willebrand factor in plasma. The likelihood of responding correlates with the initial VWF:Ag level, such that patients with very low VWF usually do not respond well (Sadler et al., 2006).

Mapping

Some small families with VWD type 1 have been observed to cosegregate with polymorphic markers at the VWF locus (Cumming et al., 1992; Inbal et al., 1992), whereas lack of association between the type 1 phenotype and intragenic markers of the VWF gene on chromosome 12p13 have been reported in families from a population study (Castaman et al., 1999).

In a study in Spain, Casana et al. (2001) performed linkage analysis of 12 families with definite or possible type 1 VWD. One family with classic type 1 had a high lod score (maximum lod = 5.28 at theta = 0.00). Four families with fully penetrant disease had a total lod score of 10.68. In 2 families, linkage was rejected, and 3 families did not show conclusive evidence of linkage.

Linkage disequilibrium mapping may be useful in genomic regions of less than 1 cM where the number of informative meioses needed to detect recombinant individuals within pedigrees is exceptionally high. Its utility for refining target areas for candidate disease genes before initiating chromosomal walks and cloning experiments is based on an understanding of the relationship between linkage disequilibrium and physical distance. To address this matter, Watkins et al. (1994) characterized linkage disequilibrium in a 144-kb region of the VWF gene in 60 CEPH and 12 von Willebrand families. The analysis revealed a general trend in which linkage disequilibrium dissipates more rapidly with physical distance in telomeric regions than in centromeric regions. This trend was consistent with higher recombination rates near telomeres.

Molecular Genetics

Sadler and Ginsburg (1993) reported on a database of polymorphisms in the VWF gene and pseudogene; Ginsburg and Sadler (1993) reported on a database of point mutations, insertions, and deletions.

Eikenboom et al. (1996) described a family in the Netherlands in which 3 affected members with type 1 von Willebrand disease and VWF levels 10 to 15% of normal were heterozygous for a mutation in the VWF gene (C1149R; 613160.0028). The mutation resulted in a decrease in the secretion of coexpressed normal VWF, and the mutation was proposed to cause intracellular retention of pro-VWF heterodimers.

'Vicenza' Variant of Von Willebrand Factor

In affected members of 7 Italian families and in 1 German patient with von Willebrand disease 'Vicenza,' Schneppenheim et al. (2000) identified a heterozygous R1205H mutation in the VWF gene (613160.0027). Haplotype identity, with minor deviations in 1 Italian family, suggested a common but not very recent genetic origin of R1205H. Von Willebrand disease 'Vicenza' was originally described in patients living in the region of Vicenza in Italy (Mannucci et al., 1988). A number of additional families were identified in Germany by Zieger et al. (1997). The phenotype was characterized by these groups as showing autosomal dominant inheritance and low levels of VWF antigen in the presence of high molecular weight and ultra high molecular weight multimers, so-called 'supranormal' multimers, similar to those seen in normal plasma after infusion of desmopressin. Randi et al. (1993) had demonstrated genetic linkage between the VWF gene and the 'Vicenza variant' of VWD type 1, suggesting that the disorder is due to a mutation in the VWF gene that results in an abnormal VWF molecule that is processed to a lesser extent than a normal VWF.

Casonato et al. (2002) identified 4 additional families with the R1205H variant. Individuals with this variant showed a mild bleeding tendency and significant decrease in plasma VWF antigen and ristocetin cofactor activity, but normal platelet VWF levels. Larger than normal VWF multimers were also observed. However, VWF multimers disappeared rapidly from the circulation after desmopressin, indicating reduced survival of the mutant protein. Since ristocetin-induced platelet aggregation was normal, Casonato et al. (2002) attributed the phenotype to reduced survival of normally synthesized VWF, which is consistent with type 1 VWF.

Cumming et al. (2006) identified the Vicenza variant in 4 (12.5%) of 32 UK patients with type 1 VWD. The R1205H mutation was highly penetrant and consistently associated with moderate to severe type 1 disease. VWF multimer studies did not show the presence of ultralarge multimers in any affected individuals; the authors thus classified the Vicenza variant to be a type 1 quantitative defect, rather than a type 2M qualitative defect as had been suggested by Castaman et al. (2002). Three of the 4 families reported by Cumming et al. (2006) shared the same haplotype, suggesting a common origin of the mutation.

Susceptibility Alleles

In 10 Canadian families and 2 families from the U. K. with type 1 VWD, O'Brien et al. (2003) identified a heterozygous Y1584C (613160.0029) substitution in the VWF gene.

Bowen and Collins (2004) described a patient with type 1 von Willebrand disease in whom the von Willebrand factor showed increased susceptibility to proteolysis by ADAMTS13 (604134). Investigation of additional family members indicated that increased susceptibility was heritable, but it did not track uniquely with type 1 VWD. Sequence analysis showed that increased susceptibility to proteolysis tracked with the Y1584C substitution. A prospective study of 200 individuals yielded 2 Y1584C heterozygotes; for both, plasma VWF showed increased susceptibility to proteolysis.

Cumming et al. (2006) identified heterozygosity for the Y1584C variant in 8 (25%) of 32 U.K. families and in 19 (17%) of 119 related individuals with type 1 VWD. Eighteen (95%) of the 19 individuals were blood group O. Heterozygosity for Y1584C segregated with VWD is 3 families, did not segregate with VWD in 4 families, and showed equivocal results in 2 families. Cumming et al. (2006) concluded that Y1594C is a polymorphism that is frequently associated with type 1 VWD, but shows incomplete penetrance and does not consistently segregate with the disease. The association with blood group type O may be related to the fact that both blood group O and Y1584C are associated with increased proteolysis of VWF by ADAMTS13 (604134).

Inheritance

Mannucci (2004) stated that type 1 VWD is typically transmitted as an autosomal dominant trait.

Cumming et al. (2006) concluded that type 1 VWD is best considered a complex multifactorial disorder, with interrelating genetic and environmental components. In a study of 32 UK families with type 1 VWD, Cumming et al. (2006) found that putative mutations did not always segregate with the disease, indicating incomplete penetrance, and that 47% of index cases did not have VWF mutations. In addition, linkage analysis of affected families does not always show linkage to the VWF locus, suggesting that additional loci may be involved.

Population Genetics

Miller et al. (1986) ascertained and studied 5 families with both VWD and hemophilia A. This suggested to them that VWD is a rather frequent disorder. They proposed 1.4% as the frequency of VWD heterozygotes.

A frequency of 2.5 to 5.0% was estimated for VWD heterozygotes in Sweden (Bowie, 1984).

Sutherland et al. (2009) identified a recurrent 8.6-kb deletion of exons 4 and 5 of the VWF gene (613160.0038) in Caucasian British patients with VWD type 1 and VWD type 3. The deletion was not found in VWD patients of Asian origin, and haplotype analysis confirmed a founder effect in the white British population. The mutation was unusual in that the truncated VWF protein produced had a dominant-negative effect on expression of the second allele.

History

Von Willebrand (1926, 1931) discovered a hemorrhagic condition in persons living on the Aland Islands in the Sea of Bothnia between Sweden and Finland and called it 'pseudohemophilia.' (See 300600 for another Aland Island disease.) The main difference from classic hemophilia was prolonged bleeding time. Major clinical problems were gastrointestinal, urinary, and uterine bleeding; hemarthroses were rare, but present. Von Willebrand and Jurgens (1933), using the capillary thrombometer, suggested the designation 'constitutional thrombopathy.' Nyman et al. (1981) followed up on the kindred originally reported by von Willebrand (1926). The relatively severe phenotype suggesting that some members of the family, which showed evidence of consanguinity, may have had VWD type 3 (277480), which is transmitted as an autosomal recessive disorder. Zhang et al. (1993) stated that the original family reported by von Willebrand (1926) carried a common 1-bp deletion in the VWF gene (613160.0021).

Alexander and Goldstein (1953) discovered low antihemophilic globulin (AHG, or factor VIII; 300841) in this disorder. Thereafter the condition became known as 'vascular hemophilia.' In the next 10 years, the main developments were the demonstrations that (1) the platelet is intrinsically normal but has reduced adhesiveness because of factor VIII deficiency, and (2) plasma from persons with classic hemophilia will correct both the vascular defect and the factor VIII deficiency.

Goldin et al. (1980) reported a hint of linkage with glutamate-pyruvate transaminase (GPT1; 138200), but Verp et al. (1982) found no evidence of linkage to GPT1.

Pickering et al. (1981) found mitral valve prolapse in 9 of 15 patients (60%) with von Willebrand disease and 4 of 30 sex- and age-matched healthy controls (13.3%). They suggested that von Willebrand disease is 'a mesenchymal dysplasia that resembles the heritable disorders of connective tissue.'

Animal Model

Von Willebrand disease has been identified in a number of mammalian species. The disease is autosomal recessive in the pig (Fass et al., 1979). Bahou et al. (1988) studied a family of pigs with the porcine form of VWD. Using a RFLP lying within the porcine VWF gene, they showed tight linkage with the disease (lod score of 5.3 with no crossovers).

Bowie et al. (1986) reported that bone marrow transplantation in a pig with severe VWD caused only partial correction of the bleeding problem. They concluded that the 'plasmatic compartment is only minimally replenished by the VWF from platelets and megakaryocytes.' Both endothelial cells and megakaryocytes synthesize VWF.

Sweeney et al. (1990) described a mouse model for human type I VWD. The affected mice showed prolonged bleeding time, normal VWF multimer distribution, autosomal dominant inheritance, and proportionately decreased plasma VWF antigen, ristocetin cofactor, and factor VIII activities. Genetic linkage analysis indicated that murine VWD is caused by a defect at a novel genetic locus, distinct from the murine VWF gene. Mohlke et al. (1996) extended these studies by localizing the major gene modifying VWF cells in the inbred mouse strain originally studied by Sweeney et al. (1990). A novel locus accounting for 63% of the total variants in VWF level was mapped to distal mouse chromosome 11, which is distinct from the murine Vwf locus on chromosome 6. They designated this locus Mvwf for 'modifier of VWF.' A single dominant gene accounting for the low VWF phenotype of the original strain studied (RIIIS/J) was demonstrated in crosses with several other strains. The pattern of inheritance suggested a gain-of-function mutation in a unique component of VWF biosynthesis or processing. Mohlke et al. (1996) commented that characterization of the human homolog of Mvwf may have relevance for a subset of type 1 VWD cases and may define an important genetic factor modifying penetrance and expression of mutations at the VWF locus. Mohlke et al. (1999) found that Mvwf, the gene responsible for modifying plasma levels of Vwf in a strain of mice, was Galgt2 (B4GALNT2; 111730). They showed that the mutation causing deficiency of Vwf changed expression of the gene from gut specific to the vascular endothelium. Ginsburg (1999) expressed the opinion that the same alteration is highly unlikely in humans. However, the Mvwf modifier gene effect may be a useful paradigm for understanding genetic modification of plasma VWF levels.

Denis et al. (1998) generated a mouse model for von Willebrand disease by using gene targeting. VWF-deficient mice appeared normal at birth; they were viable and fertile. Neither von Willebrand factor nor VWF-propolypeptide (von Willebrand antigen II) was detectable in plasma, platelets, or endothelial cells of the homozygous mutant mice. The mutant mice exhibited defects in hemostasis with a highly prolonged bleeding time and spontaneous bleeding events in approximately 10% of neonates. As in the human disease, the factor VIII level in these mice was reduced strongly as a result of the lack of protection provided by von Willebrand factor. Defective thrombosis in mutant mice was also evident in an in vivo model of vascular injury. In this model, the exteriorized mesentery was superfused with ferric chloride and the accumulation of fluorescently labeled platelets was observed by intravascular microscopy. Denis et al. (1998) concluded that these mice very closely mimic severe human von Willebrand disease.