Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Mar 1.
Published in final edited form as: J Thromb Haemost. 2023 Nov 30;22(3):666–675. doi: 10.1016/j.jtha.2023.11.016

The common VWF variant p.Y1584C: Detailed pathogenic examination of an enigmatic sequence change

Pamela A Christopherson 1, Nathalie Tijet 2, Sandra L Haberichter 1, Veronica H Flood 1, Justyne Ross 3, Colleen Notley 2, Orla Rawley 2, Robert R Montgomery 1; Zimmerman Project Investigators, Paula D James 4, David Lillicrap 2
PMCID: PMC10922911  NIHMSID: NIHMS1955319  PMID: 38040335

Abstract

Background

As knowledge of the human genome has advanced, so too has the recognition that interpretation of the pathogenic nature of sequence variants can be challenging. The von Willebrand factor (VWF) gene exhibits a significant degree of sequence variability and the first VWF variant associated with type 1 von Willebrand disease (VWD), c.4751 A>G, p.Y1584C, was described in 2003. However, since that time, the pathogenic nature of this variant has remained unclear, being assigned properties ranging from a risk factor to a pathogenic variant.

Objectives

We aim to provide additional evaluation on the interpretation of pathogenicity for this common VWF variant.

Patients/Methods

Fifty-eight subjects with only the p.Y1584C variant were recruited from two cohort studies (the Zimmerman Program and the Canadian Type 1 VWD study). Clinical and laboratory phenotypes were assessed.

Results

The prevalence of the p.Y1584C variant in our cohorts was 23 to 27-fold higher than in large normal population databases. Significantly more p.Y1584C subjects had an abnormal bleeding score compared to Y1584 individuals. In comparison with a group of 35 subjects without the p.Y1584C variant, subjects with the variant had lower mean VWF:Ag and VWF:RCo values and significantly higher VWFpp/VWF:Ag ratios suggestive of enhanced clearance.

Conclusions

Collectively, the results of this analysis suggest that p.Y1584C is likely pathogenic in nature, however, due to influences such as incomplete penetrance, variable expressivity and other genetic modifiers like ABO blood group, the straightforward assignment of pathogenicity to this variant is inevitably challenging.

Keywords: von Willebrand disease, von Willebrand factor, pathogenic variant, p.Y1584C

Introduction

Since the cloning of the von Willebrand factor gene (VWF) in the mid-1980s1–4 the molecular genetic landscape of the common inherited bleeding disorder von Willebrand disease (VWD) has been extensively investigated by many groups around the world. These studies have been variably successful in identifying pathogenic variants that are responsible for the various subtypes of VWD.

Most success has been achieved with the type 2 VWD subtypes for which causative VWF variants are usually found in >90% of symptomatic patients with VWF protein phenotypes consistent with types 2A, 2B, 2M and 2N disease. These genetic analyses are now frequently used to confirm the results of phenotypic studies or when definitive phenotypic tests are unavailable (eg. FVIII binding and VWF multimer analysis).5,6

Unfortunately, despite the success of molecular genetic testing for type 2 VWD, these subtypes only represent ~25% of all VWD cases, and in reality, experienced clinical coagulation laboratories can make the diagnosis of types 2A, 2B and 2M disease based on specialized but routine phenotypic assays. Thus, genetic diagnosis, while reassuring when phenotypic tests are uncertain, is usually not part of the standard VWD diagnostic process.

The two quantitative forms of VWD are the common type 1/Low VWF subtypes that represent ~75% of cases and the rare type 3 VWD that has a population prevalence of ~0.5–3 per million depending largely upon rates of consanguinity. In these VWD subtypes, the phenotypic diagnosis rests on robust documentation of low plasma levels of normally functional VWF. As VWF is an acute phase reactant and is also influenced by factors such as estrogens and physical stress, temporal variability of plasma levels is frequent and thus definitive documentation of low VWF levels will often require repeat testing.7–10

Against this background, and with the VWF sequence being available, there were initial hopes that the type and pattern of genetic variants responsible for these quantitative traits might provide a more reliable diagnostic test. However, after >15 years of investigation by many laboratories around the world this proposal remains unproven due in large part to the highly polymorphic nature of the VWF locus and to difficulties in proving pathogenicity of the many variants that are present.

In this report, we examine the subject of type 1 VWD genetic diagnosis with a focus on one missense variant, the p.Y1584C substitution that was the first type 1 variant identified, 20 years ago.11 Through the collective efforts of two large multicenter studies, the Zimmerman Program for the Molecular and Clinical Biology of VWD12,13 and the Canadian type 1 VWD Study14 58 cases with the p.Y1584C variant have been identified with accompanying detailed clinical and laboratory phenotypes. This report summarizes these findings and provides commentary on the interpretation of pathogenicity for this common VWF variant.

Materials and Methods

Patient Populations

Subjects were enrolled in the Zimmerman Program for the Molecular and Clinical Biology of VWD (Zimmerman Program) study through 10 primary and 23 secondary clinical centers across the United States. Informed consent was obtained using protocols approved by each local Internal Review Board (IRB). Index cases with a pre-existing diagnosis of VWD and their affected and unaffected family members were included. Affected family members and unaffected family members were determined either by a previous VWD diagnosis or VWF levels obtained at their baseline study visit. 244 healthy controls from the general population were enrolled through 7 primary centers in Atlanta, Detroit, Houston, Indianapolis, Milwaukee, New Orleans, and Pittsburgh. Inclusion criteria for healthy controls included being over the age of 18 and no previous diagnosis of a bleeding disorder.

In the Canadian type 1 VWD study, subjects were enrolled from 13 academic health science centres across Canada. All subjects were informed of the experimental nature of the study and gave informed consent. The study was approved by the Research Ethics Board of Queen’s University and at each of the source institutions.

Bleeding Score Determination

Bleeding symptoms were quantified using the International Society on Thrombosis and Haemostasis (ISTH-BAT) bleeding score only in subjects enrolled in the Zimmerman Program. The bleeding score questionnaire was administered to each subject by a trained coordinator, nurse, or physician where subjects were encouraged but not required to answer all questions. All bleeding symptoms up to the time of enrollment were captured. The cut-off for abnormal bleeding scores is ≥ 4 in adult males, ≥ 6 in adult females and ≥ 3 in children.15

Laboratory Testing - Phenotypic analysis

Zimmerman Program:

Citrated blood was collected at time of enrollment and processed for plasma. All testing was performed centrally at the Versiti Blood Center of Wisconsin Hemostasis Reference Laboratory which included Factor VIII activity (FVIII), VWF antigen (VWF:Ag), VWF ristocetin cofactor (VWF:RCo), VWF propeptide (VWFpp), VWF collagen binding (VWF:CB), reverse blood type (ABO) and multimer analysis. Additional confirmatory studies were done in the research laboratory at the Versiti Blood Research Institute as previously described.16 VWF:Ag was performed by enzyme-linked immunosorbent assay and VWF:RCo by automated platelet agglutination.17 Multimer distribution was assayed by quantitative gel electrophoresis.18 VWF propeptide (VWFpp) was measured to evaluate potential VWF clearance defects.19 The ABO Blood type was ascertained by reverse typing.17

Subjects (index cases and affected family members) were classified as low VWF (LVWF) if they had central laboratory results for VWF:Ag of 30–50 IU/dl or VWF:RCo 30–53 IU/dl, type 1 if they had VWF:Ag or VWF:RCo <30 IU/dl, type 1C for VWFpp/VWF:Ag ≥3 and normal multimers, or type 1H if they had historical low VWF levels but had normal levels at time of study enrollment. The unaffected family members had a normal VWF:Ag and VWF:RCo level (>50 IU/dl). The bleeding score was not used to define the affected and unaffected family members.

Canadian Type 1 VWD Study:

Whole blood samples were collected by phlebotomy in 3.2% sodium citrate (at a ratio of 9:1) from the index case. A plasma sample was collected and frozen from the index case to confirm the diagnosis of type 1 VWD in a central Laboratory. Laboratory tests for VWF:Ag and VWF:RCo were performed at the source clinic attended by the patient according to local methods. These tests were confirmed on frozen plasma samples at the Clinical Hemostasis Laboratory at Kingston General Hospital as previously described.14 The classification of type 1 VWD index case with p.Y1584C variant used the same criteria as the one used by the Zimmerman program. The ABO blood group nucleotide 261 (rs8176719) and nucleotide 703 (rs8176743) SNPs were genotyped using the 5’ nuclease assay to distinguish ABO blood group.20

Although a formal comparison of phenotype testing was not conducted, both centers (Milwaukee and Kingston) use the same test reagents and a prior (unpublished) comparison of VWF:Ag and VWF:RCo results showed good correlation.

DNA sequencing

For subjects from the Zimmerman Program, EDTA anticoagulated whole blood was collected, and DNA was isolated by the Qiagen Gentra Puregene method in the Versiti Molecular Diagnostic Laboratory. Full VWF exonic Sanger sequencing was performed, including intron/exon boundaries, at the Harvard Partners Genome Center (HPGC), Versiti Blood Center of Wisconsin or Functional Biosciences (Madison, Wisconsin). The VWF reference sequence NM_000552 was used and variant analysis was performed using Softgenetics Mutation Surveyor DNA Variant Analysis Software. Candidate VWF variants identified in the index case were confirmed in all family members using direct targeted sequencing.

For the Canadian type 1 VWD study, an EDTA blood sample was collected from the index cases and genomic DNA was isolated from leukocytes using a salt extraction method. DNA sequencing was performed as previously described by James et al., 2007.14

Formal Pathogenicity Curation

The p.Y1584C variant was subjected to formal pathogenicity curation utilizing the Guidelines of the American College of Medical Genetics and Genomics (ACMGG) and the Association for Molecular Pathology (AMG).21

Statistical analysis

Data analysis (mean, median) was completed using GraphPad Prism 9.5.1. The Mann-Whitney U-test was performed for the comparison of two groups and Kruskal-Wallis test for multiple comparison (three groups or more).

Results

Families and Patients

Zimmerman Program:

A total of 2901 patients enrolled in the Zimmerman Program belonging to 743 families and collected from 33 clinical centres were assessed (Figure 1). From these 743 families, 32 families had an index case with the p.Y1584C variant (2.2% of alleles). Twelve families had index cases with at least one other pathogenic variant and to eliminate the influence of these other variants, twenty families with their index cases only having the p.Y1584C variant were examined. Lastly, five families were removed from the final analysis due to their index case diagnostic analysis: one had a type 2A VWD profile and four type 1H (historical) phenotype. Thus, a total of 44 p.Y1584C subjects (15 index cases, 12 affected family members and 17 unaffected family members) belonging to 15 families and collected from 8 clinical centres were selected for further study.

Figure 1: Flow chart of selection of subjects enrolled in Zimmerman Program for Molecular and Clinical Biology of VWD (ZPMCB-VWD) and Canadian type 1 VWD study.

Figure 1:

Open in a new tab

Diagram showing subjects who met the criteria for inclusion in, or exclusion from the study.

Of the 44 subjects (p.Y1584C subjects) who had only the p.Y1584C variant, 20 were diagnosed as Low VWF, 7 type 1 VWD and 17 subjects were unaffected family members. Forty-one subjects showed normal multimer profiles. The multimer interpretation for 3 subjects was unclear. Nevertheless, as two were diagnosed as Low VWF and the other one is an unaffected family member, these 3 subjects were maintained in the study. Twenty-five of these subjects (57%) were female and 19 (43%) were male. The mean age was 27 years (ranging from <1 to 64 years). Nineteen subjects were under the age of 18 years at the time of their enrollment. All the subjects recorded their race origin as white and four recorded their ethnic origin as Hispanic or Latino.

To analyse the effect of the p.Y1584C variant on VWF:Ag level, VWF:RCo activity and VWF clearance, 35 subjects (p.Y1584 subjects) with no sequence variant belonging to the 15 families selected were included in the study (Figure 1). Twenty-one (60%) were female, 14 (40%) were male, with a mean age of 25 years (ranging from <1 to 59 years). All recorded their race as white; one recorded their ethnic origin as Hispanic or Latino and one as unknown.

Canadian Type 1 VWD Study:

Similar criteria of selection were used for the Canadian type 1 VWD study (Figure 1). One hundred and twenty-three families from 13 clinical centres were examined.14 Eighteen index cases had the variant p.Y1584C, with fourteen of these cases having only this variant. All these cases were classified as type 1 VWD and were collected from 4 clinical centres. Four were males (28.5%) and 10 females (71.5%). The mean age was 30 years (ranging from 3 to 46 years). All recorded their race as white.

Combined Study Population:

Collectively, the study population for this report involved 58 p.Y1584C subjects from 29 families with only the p.Y1584C variant and recruited from 12 clinical centers. Thirty-five p.Y1584 subjects without the p.Y1584C variant from 15 of the recruited families as well as 244 healthy controls were analysed for comparisons of VWF:Ag, VWF:RCo and VWFpp/VWF:Ag ratios.

Prevalence of p.Y1584C

p.Y1584C is shown to be present in healthy populations at a heterozygote prevalence ranging from 0.08 to 0.27% (Table 1). The Canadian type 1 VWD Study identified p.Y1584C in 18 index cases from 123 families showing a prevalence of 7.3%. In the Zimmerman Program, VWF gene sequencing of 743 index cases identified p.Y1584C in 32 index cases (2.2%). Thirty-one index cases were heterozygotes, and one was found to be a homozygote. However, the subjects enrolled in the Zimmerman Program have been diagnosed with a wide range of VWD subtypes including low VWF, type1 VWD, type 1H VWD, type 1C VWD, type 2 VWD and type 3 VWD. If only the index cases with type 1 VWD and low VWF are considered, the prevalence of the p.Y1584C variant in the Zimmerman Program is 6.2%, a similar frequency compared to the Canadian type 1 VWD Study. Therefore, the p.Y1584C prevalence in the Zimmerman Program (type 1 and Low VWF cases) and the Canadian type 1 VWD Study is at least 23-fold and 27-fold higher respectively compared to that recorded in large healthy population databases.

Table 1.

p.Y1584C frequency (%) in the general population

Population gnomAD genome (v3.1.2) gnomAD exome (v2.1.1) 1000 genomes UK Biobank
Total 0.23 0.27 0.08 na
European Non-Finnish 0.39 0.40 na 0.43
African/African American 0.041 0.056 na na
South Asian 0.041 0.042 na na
Open in a new tab

Review of the gnomAD database (v2.1.1) has documented 4 homozygotes in 141,243 subjects (prevalence 0.000028). In the Zimmerman Program, one homozygous p.Y1584C subject with blood group A and VWF:Ag of 69 IU/dl was identified. Furthermore, in the original report of this variant published in 2003,11 one homozygous p.Y1584C case from United Kingdom, was documented. This case was blood group O and had a significantly lower VWF:Ag (24 IU/dl).

ABO Group Distribution

ABO blood group frequencies differ between p.Y1584C and p.Y1584 subjects (Table 2) and also with the general population in North America. The frequency of group O is 56.9% in p.Y1584C subjects compared to 40% in p.Y1584 individuals, and in contrast to the Canadian and USA blood group O frequencies of 46% and 44%, respectively. Blood group A frequency was 39.6% in p.Y1584C individuals and 51.4% in p.Y1584 subjects (group A frequency in USA and Canada is 42%).

Table 2.

ABO blood group frequencies for p.Y1584C and p.Y1584 subjects

ABO blood p.Y1584C p.Y1584
group Total No % Total No %
O 33 56.9 14 40
A 23 39.7 18 51.4
AB 1 1.7 0 0
B 1 1.7 2 5.7
Unknown 1* 2.9
Open in a new tab
*

One p.Y1584 subject was too young to determine the ABO blood type.

Bleeding Score

ISTH bleeding scores were only available for Zimmerman Program subjects (Figure 2). Using appropriate age and gender cutoffs for the ISTH-BAT bleeding score, 36% of Zimmerman subjects with p.Y1584C had an abnormal bleeding score whereas only 20% of p.Y1584 subjects had an abnormal score. Of note, the median bleeding score in healthy controls (1; IQR 2, range 0–14) is no different (p=0.0599) to subjects with p.Y1584 (1, IQR 3, range 0–9).

Figure 2: ISTH Bleeding Scores in the Zimmerman Program subjects.

Figure 2:

Open in a new tab

ISTH BAT bleeding scores (BS) for Y1584C subjects (N=33), Y1584 subjects (N=35) and healthy controls (HC; N=244).

The median bleeding score (BS) was not significantly higher based on Zimmerman data alone (Figure 2). Bleeding scores were missing for 11 p.Y1584C subjects including 7 females and 4 males.

The median BS for p.Y1584C and p.Y1584 females is not significantly different. The same results were found for males. However, 40% of p.Y1584C females and 32% of p.Y1584C males showed an abnormal BS compared to only 29% of p.Y1584 females and 14% of p.Y1584 males.

No significant difference in bleeding scores was noticed between p.Y1584C group O (Median 3, IQR 4.5) and p.Y1584C non group O (median 2, IQR 5).

VWF:Ag and VWF:RCo Plasma Levels

The analysis of VWF:Ag and VWF:RCo levels showed significant differences between p.Y1584C and p.Y1584 subjects (Figure 3, A and C). The mean VWF:Ag and VWF:RCo levels were significantly lower (p<0.0001) for p.Y1584C subjects (VWF:Ag 57 IU/dl ; VWF:RCo 53 IU/dl) compared to p.Y1584 subjects (VWF:Ag 83.1 IU/dl ; VWF:RCo 82.3 IU/dl). Nevertheless, these values were still within the lower range of normal values (VWF:Ag >50 and VWF:RCo ≥53 IU/dl) even for p.Y1584C subjects. As well, the mean VWF:Ag and VWF:RCo levels for p.Y1584C females and males were significantly lower compared to p.Y1584 females and males respectively (data not shown). There was no significant difference in VWF:Ag and VWF:RCo between p.Y1584C males and p.Y1584C females. VWF:Ag levels ranged from 7–144 IU/dl for p.Y1584C subjects and 27–180 IU/dl for p.Y1584 subjects. VWF:RCo values ranged from <10–164 IU/dl in p.Y1584C and 29–170 IU/dl in p.Y1584 subjects respectively.

Figure 3: VWF:Ag and VWF:RCo levels in p.Y1584C and p.Y1584 subjects and in blood group O and non-O subjects.

Figure 3:

Open in a new tab

Panels A & B: VWF:Ag levels. Panels C & D VWF:RCo levels. Significant decrease of VWF:Ag (A) and VWF:RCo (C) levels is observed in Y1484C subjects (n=58) compared to p.Y1584 subjects (n=35). P value using the Mann-Whitney test. VWF:Ag (B) and VWF:RCo (D) is significantly lower for group O p.Y1584C subjects (n=33) compared to non-group O p.Y1584C (n=25) and to group O p.Y1584 (n=14). No significant difference was observed for the p.Y1584 subjects between blood group O and non-group O (P>0.9999). P value using Kruskal-Wallis test).

Thirty five (60%) p.Y1584C subjects had VWF:Ag ≤50 IU/dl compared to only 5 (14%) for p.Y1584 subjects with a similar VWF:RCo range (<52 IU/dl) for all subjects. Twelve p.Y1584C subjects with VWF:Ag and VWF:RCo levels ≤50 IU/dl and six with normal VWF:Ag and VWF:RCo level have an abnormal BS.

The mean VWF:Ag and VWF:RCo levels were significantly lower (p=0.0008 and p=0.0013 respectively) for ABO blood group O p.Y1584C subjects (VWF:Ag 43 IU/dl ; VWF:RCo 40 IU/dl) compared to non-group O (VWF:Ag 75 IU/dl ; VWF:RCo 71.1 IU/dl) (Figure 3, B and D). No significant difference was observed for the p.Y1584 subjects between ABO blood group O and non-group O. This data showed a larger decrease of VWF:Ag and VWF:RCo levels for p.Y1584C subjects and ABO group O subjects.

Because the p.Y1584C variant has been shown to be in phase with the c.5312-19A>C variant in VWF intron 30 at a frequency of 10%,11 and was not found in the healthy population, we investigated the combination of these two VWF variants on VWF levels. Sequencing for the c.5312-19A>C in VWF intron 30 was performed only in the index cases in the Zimmerman Program and the Canadian type 1 VWD study and not for any additional family members. From 29 index cases sequenced, this variant was found in phase with p.Y1584C in 12 index cases (41%). However the mean VWF:Ag and VWF:RCo levels showed no significant difference for the subjects with the combination of these two variants compared to the subjects with only the p.Y1584C variant (data not shown).

VWF Clearance

VWF propeptide levels were only available for subjects enrolled in the Zimmerman Program.

The mean VWFpp/VWF:Ag ratio for p.Y1584C and p.Y1584 subjects was significantly lower compared to confirmed VWD type 1C subjects, but significantly higher to the healthy controls in this study (Figure 4A).

Figure 4: VWF clearance in type 1C, p.Y1584C, p.Y1584 Zimmerman Program subjects and in healthy controls and in blood group O and non-group O subjects.

Figure 4:

Open in a new tab

VWFpp/VWF:Ag ratio was measured as a marker of VWF clearance. Panels (A) and (B): type 1C (n=125); p.Y1584C (n=39); p.Y1584 (n=33); HC (n=244). Panel (C): p.Y1584C O (n=21), p.Y1584C non-O (n=18), p.Y1584 (n=13), p.Y1584 non-O (n=19). P value using the Mann-Whitney test and Kruskal-Wallis test. HC: Healthy controls.

Furthermore, analysis of the VWFpp/VWF:Ag ratio indicates that VWF clearance is increased in p.Y1584C subjects compared to p.Y1584 subjects (Figure 4, A and B). The mean VWFpp/VWF:Ag ratio was significantly higher (p=0.0003) for p.Y1584C individuals compared to p.Y1584 subjects although it was still within the normal range (VWFpp/VWF:Ag ratio 1.64 for p.Y1584C and 1.25 for p.Y1584). Nine p.Y1584C and two p.Y1584 subjects had a ratio VWFpp/VWF:Ag >2 (range from 2.02 to 3.1).

The VWFpp/VWF:Ag ratio was significantly higher in blood group O than in blood group non-O for p.Y1584C (p=0.0.0105) (Figure 4C). However, in blood group O the VWFpp/VWF:Ag ratio was not significantly higher for p.Y1584C compared to p.Y1584 subjects (p=0.0967). Additionally, data showed no significant difference for blood group non-O between p.Y1584C and p.Y1584 subjects. This data is consistent with the faster clearance of VWF for group O p.Y1584C subjects compared to blood group non-O.

Formal Pathogenicity curation of p.Y1584C

Curation of the p.Y1584C variant using the rules derived from the formal ACMG/AMP guidelines for variant analysis21 concluded that the variant was Likely Pathogenic.

Discussion

The p.Y1584C variant (exon 28, c.4751 A>G), located within the A2 domain of VWF was first identified in 14% of index cases during the Canadian type 1 VWD study with an associated founder polymorphic haplotype11,14 Currently, the pathogenic nature of this variant remains uncertain. Thus, in the ClinVar database the variant is listed as showing “conflicting interpretations of pathogenicity” with reports as a “risk factor”, Pathogenic variant (four reports), Likely Pathogenic (seven reports) and Variant of Uncertain Significance (three reports) (https://www.ncbi.nlm.nih.gov/clinvar).

The p.Y1584C variant is frequently associated with a mild type 1 VWD phenotype and shows incomplete penetrance and variable expressivity, as this variant can be present in both clinically affected and unaffected individuals in the same family. Further confirmation of the enrichment of this variant allele in VWD cases has been documented in several other studies. The p.Y1584C variant was found in 8% of index cases in the European Molecular and Clinical Markers for Diagnosis and Management (MCMDM)-1 VWD study,22 3.7% of type 1 VWD cases in Spain,23 27% of index cases in the UK Haemophilia Centre Doctors’ Organization study24 and 2.6% index cases from the Bonn Haemophilia Center and different regions in Germany.25 Most recently, a study in Nigeria has reported the presence of this variant in 7% of index cases.26 In marked contrast, the review of multiple reference databases (gnomAD, 1000 genomes, UK BioBank) has identified this variant in the general population at a prevalence ranging from 0.08% to 0.27%, with the highest prevalence of 0.43% in the European Non-Finnish population (Table 1).

The p.Y1584C variant is associated with a decrease of plasma VWF antigen (VWF:Ag), VWF collagen binding activity (VWF:CB), and ristocetin cofactor activity (VWF:RCo).27 A larger decrease of the VWF antigen was observed for ABO blood group O individuals.28 Expression studies of the mutant protein in which a new solvent-exposed thiol is produced, showed a reduction in VWF synthesis, an increase in intracellular retention of VWF protein and an enhanced susceptibility to ADAMTS13-mediated proteolysis.11,29 In addition, data for the p.Y1584C variant and ABO blood group O supports evidence that these two factors combine to influence VWF clearance.28 In the VWF knockout mouse model using recombinant VWF protein infusion and hydrodynamic delivery of wild type and p.Y1584C VWF cDNA, p.Y1584C showed no change in protein clearance, decreased VWF antigen levels and a mild reduction in high molecular weight multimers.30

The p.Y1584C variant, present in the general population at a very low frequency (0.08% to 0.27%), showed a similar prevalence in males and females (gnomAD Database, data not shown). However, the prevalence of the variant is different in populations of varying ethnicity. Thus, the frequency is 10 times higher in the European Non-Finnish population than in the South Asian and African/African American population (gnomAD Database, Table 1) and it is not present in East Asian and middle Eastern population (gnomAD Database, data not shown). These observations combined with the original documentation of a founder VWF haplotype associated the variant,11 suggest a common Northern European origin for p.Y1584C.

The comparison of the p.Y1584C variant frequency among populations of VWD patients is relatively limited. Most of the studies describing the p.Y1584C variant in VWD cohorts have been reported from European and North American countries in a majority white population.11,14,22,24,25 Only one study reported the presence of p.Y1584C in VWD cases in an African country.26 In the combined study population reported here, p.Y1584C as the only variant was present in 58 subjects, 100% recorded their race origin as white and 93% their ethnic origin as non-Hispanic or non-Latino. Thus, in this cohort, the African American and Asian populations represent only 5% and 2% respectively and as a result, no significant conclusions can be drawn about the distribution of p.Y1584C in VWD patients of different ethnicities based on all the available data.

In addition to the various ClinVar database assertions alluded to above, (https://www.ncbi.nlm.nih.gov/clinvar) in silico evaluation of p.Y1584C’s pathogenicity with the CADD (Combined Annotation Dependent Depletion) Score31 predicts that this variant is deleterious (score of 26). Furthermore, evaluation of pathogenicity with the very recently released AlphaMissense AI program,32 that incorporates structural biology and species conservation details from the AlphaFold program,33 also assigns p.Y1584C as a likely pathogenic variant (metric 0.5865).

In our combination cohort study, p.Y1584C is associated with a decrease of VWF:Ag and VWF:RCo levels and this decrease is accentuated in blood group O subjects. Additionally, VWF clearance is mildly accelerated in subjects with the p.Y1584C variant and for group O subjects. Collectively, these observations suggest that p.Y1584C results in lower plasma VWF levels due to a combination of biosynthetic and survival defects and that these abnormal phenotypes are the result of adverse interactions between the p.Y1584C and blood group O variants. Of note, of 33 blood group O subjects with only the p.Y1584C variant, only 13% are not clinically affected (ie. normal BS and VWF indices) confirming the ABO blood group influence on the phenotype, and also that this variant exhibits incomplete penetrance. The influence of ABO blood group and incomplete penetrance may also explain, at least in part, the phenotypic discrepancy documented in the two p.Y1584C homozygotes in this report, one with a VWF:Ag of 69 IU/dl (blood group A) and one with a VWF:Ag of 24 IU/dl (blood group O).

Interestingly, and as a complement to other lines of evidence, the p.Y1584C variant was recently identified as a significant influence in a genome wide association study for iron homeostasis.34 However, the association with iron deficiency was only observed in pre-menopausal females, and the reasonable assumption is that this association relates to increased menstrual and post-partum bleeding in women with the variant and with lower plasma VWF levels.

In conclusion, despite having identified this variant 20 years ago, the pathogenic classification of p.Y1584C remains controversial. Nevertheless, and keeping in mind its associated incomplete penetrance, variable expressivity and strong interaction with ABO blood group, we suggest that there is strong epidemiologic evidence that the variant is at the very least a marker of low plasma VWF levels. Furthermore, a range of orthogonal investigations have also documented abnormalities of VWF biosynthesis and clearance in association with the variant, suggesting that this is indeed likely a pathogenic variant. Most importantly, interpretation of the pathogenicity assessment for p.Y1584C, even with this relatively large study population and comprehensive phenotypic evaluation, highlights the challenges faced in large-scale genomic projects to correctly assign pathogenicity. These assessments must be made with methodologic rigor and incorporation of all available data to fully support the assignments.

Supplementary Material

1

Acknowledgements

This analysis was supported by funds from the Zimmerman Program for Molecular and Clinical Biology of von Willebrand Disease by The National Institutes of Health Program Project Grant P01HL081588 and P01HL144457 and a Canadian Institutes of Health Research Foundation Grant (FDN 154285).

Disclosures of Potential Conflicts

DL receives research support from Bayer, Biomarin, CSL-Behring and Sanofi and acts as an advisor to Biomarin, CSL-Behring, Pfizer and Sanofi. PJ receives research support from Bayer and acts as a Consultant to Band/Guardian Therapeutics and Star/Vega Therapeutics. The other authors have no relevant conflicts of interest to declare.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  • 1.Sadler JE, Shelton-Inloes BB, Sorace JM, et al. Cloning and characterization of two cDNAs coding for human von Willebrand factor. Proc. Natl. Acad. Sci. U. S. A 1985;82(19):6394–6398. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Ginsburg D, Handin RI, Bonthron DT, et al. Human von Willebrand factor (vWF): isolation of complementary DNA (cDNA) clones and chromosomal localization. Science. 1985;228(4706):1401–1406. [DOI] [PubMed] [Google Scholar]
  • 3.Verweij CL, de Vries CJ m, Distel B, et al. Construction of cDNA coding for human von willebrand factor using antibody probes for colony-screening and mapping of the chromosomal gene. Nucleic Acids Res. 1985;13(13):4699–4717. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Lynch DC, Zimmerman TS, Collins CJ, et al. Molecular cloning of cDNA for human von Willebrand factor: authentication by a new method. Cell. 1985;41(1):49–56. [DOI] [PubMed] [Google Scholar]
  • 5.Itzhar-Baikian N, Boisseau P, Joly B, Veyradier A. Updated overview on von Willebrand disease: focus on the interest of genotyping. Expert Rev. Hematol 2019;12(12):1023–1036. [DOI] [PubMed] [Google Scholar]
  • 6.Batlle J, Pérez-Rodríguez A, Corrales I, et al. Update on Molecular Testing in von Willebrand Disease. Semin. Thromb. Hemost 2019;45(7):708–719. [DOI] [PubMed] [Google Scholar]
  • 7.Blombäck M, Eneroth P, Andersson O, Anvret M. On laboratory problems in diagnosing mild von Willebrand’s disease. Am. J. Hematol 1992;40(2):117–120. [DOI] [PubMed] [Google Scholar]
  • 8.Abildgaard CF, Suzuki Z, Harrison J, Jefcoat K, Zimmerman TS. Serial studies in von Willebrand’s disease: variability versus “variants.” Blood. 1980;56(4):712–716. [PubMed] [Google Scholar]
  • 9.Nichols WL, Hultin MB, James AH, et al. von Willebrand disease (VWD): evidence-based diagnosis and management guidelines, the National Heart, Lung, and Blood Institute (NHLBI) Expert Panel report (USA). Haemophilia. 2008;14(2):171–232. [DOI] [PubMed] [Google Scholar]
  • 10.James PD, Connell NT, Ameer B, et al. ASH ISTH NHF WFH 2021 guidelines on the diagnosis of von Willebrand disease. Blood Adv. 2021;5(1):280–300. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.O’Brien LA, James PD, Othman M, et al. Founder von Willebrand factor haplotype associated with type 1 von Willebrand disease. Blood. 2003;102(2):549–557. [DOI] [PubMed] [Google Scholar]
  • 12.Flood VH, Gill JC, Friedman KD, et al. Von Willebrand disease in the United States: a perspective from Wisconsin. Semin. Thromb. Hemost 2011;37(5):528–534. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Flood VH, Abshire TC, Christopherson PA, et al. Von Willebrand disease in the United States: perspective from the Zimmerman program. Ann. Blood 2018;3:7–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.James PD, Notley C, Hegadorn C, et al. The mutational spectrum of type 1 von Willebrand disease: Results from a Canadian cohort study. Blood. 2007;109(1):145–154. [DOI] [PubMed] [Google Scholar]
  • 15.Elbatarny M, Mollah S, Grabell J, et al. Normal range of bleeding scores for the ISTH-BAT: adult and pediatric data from the merging project. Haemophilia. 2014;20(6):831–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Flood VH, Christopherson PA, Gill JC, et al. Clinical and laboratory variability in a cohort of patients diagnosed with type 1 VWD in the United States. Blood. 2016;127(20):2481–2488. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Flood VH, Gill JC, Morateck PA, et al. Common VWF exon 28 polymorphisms in African Americans affecting the VWF activity assay by ristocetin cofactor. Blood. 2010;116(2):280–286. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Jacobi PM, Gill JC, Flood VH, et al. Intersection of mechanisms of type 2A VWD through defects in VWF multimerization, secretion, ADAMTS-13 susceptibility, and regulated storage. Blood. 2012;119(19):4543–4553. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Haberichter SL, Balistreri M, Christopherson P, et al. Assay of the von Willebrand factor (VWF) propeptide to identify patients with type 1 von Willebrand disease with decreased VWF survival. Blood. 2006;108(10):3344–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Olsson ML, Chester MA. Polymorphism and recombination events at the ABO locus: a major challenge for genomic ABO blood grouping strategies. Transfus. Med 2001;11(4):295–313. [DOI] [PubMed] [Google Scholar]
  • 21.Richards S, Aziz N, Bale S, et al. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet. Med 2015;17(5):405–424. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Goodeve A, Eikenboom J, Castaman G, et al. Phenotype and genotype of a cohort of families historically diagnosed with type 1 von Willebrand disease in the European study, Molecular and Clinical Markers for the Diagnosis and Management of Type 1 von Willebrand Disease (MCMDM-1VWD). Blood. 2007;109(1):112–121. [DOI] [PubMed] [Google Scholar]
  • 23.Borràs N, Batlle J, Pérez-Rodríguez A, et al. Molecular and clinical profile of von Willebrand disease in Spain (PCM-EVW-ES): comprehensive genetic analysis by next-generation sequencing of 480 patients. Haematologica. 2017;102(12):2005–2014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Bowen DJ, Collins PW, Lester W, et al. The prevalence of the cysteine1584 variant of von Willebrand factor is increased in type 1 von Willebrand disease: co-segregation with increased susceptibility to ADAMTS13 proteolysis but not clinical phenotype. Br. J. Haematol 2005;128(6):830–836. [DOI] [PubMed] [Google Scholar]
  • 25.Yadegari H, Driesen J, Pavlova A, et al. Mutation distribution in the von Willebrand factor gene related to the different von Willebrand disease (VWD) types in a cohort of VWD patients. Thromb. Haemost 2012;108(4):662–671. [DOI] [PubMed] [Google Scholar]
  • 26.Ezigbo E, Ukaejiofo E, Nwagha T. Molecular characterization of exon 28 of von Willebrand’s factor gene in Nigerian population. Niger. J. Clin. Pract 2017;20(2):235–238. [DOI] [PubMed] [Google Scholar]
  • 27.Davies JA, Collins PW, Hathaway LS, Bowen DJ. Effect of von Willebrand factor Y/C1584 on in vivo protein level and function and interaction with ABO blood group. Blood. 2007;109(7):2840–2846. [DOI] [PubMed] [Google Scholar]
  • 28.Davies JA, Collins PW, Hathaway LS, Bowen DJ. von Willebrand factor: Evidence for variable clearance in vivo according to Y/C1584 phenotype and ABO blood group. J. Thromb. Haemost 2008;6(1):97–103. [DOI] [PubMed] [Google Scholar]
  • 29.Bowen DJ, Collins PW. An amino acid polymorphism in von Willebrand factor correlates with increased susceptibility to proteolysis by ADAMTS13. Blood. 2004;103(3):941–947. [DOI] [PubMed] [Google Scholar]
  • 30.Pruss CM, Golder M, Bryant A, et al. Pathologic mechanisms of type 1 VWD mutations R1205H and Y1584C through in vitro and in vivo mouse models. Blood. 2011;117(16):4358–4366. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Rentzsch P, Witten D, Cooper GM, Shendure J, Kircher M. CADD: predicting the deleteriousness of variants throughout the human genome. Nucleic Acids Res. 2019;47(D1):D886–D894. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Cheng J, Novati G, Pan J, et al. Accurate proteome-wide missense variant effect prediction with AlphaMissense. Science. 2023;381(6664):eadg7492. [DOI] [PubMed] [Google Scholar]
  • 33.Jumper J, Evans R, Pritzel A, et al. Highly accurate protein structure prediction with AlphaFold. Nature. 2021;596(7873):583–589. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Bell S, Rigas AS, Magnusson MK, et al. A genome-wide meta-analysis yields 46 new loci associating with biomarkers of iron homeostasis. Commun. Biol 2021;4(1):. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS1955319-supplement-1.docx (43.3KB, docx)

RESOURCES