VWF propeptide and ratios between VWF, VWF propeptide, and FVIII in the characterization of type 1 von Willebrand disease

Key Points

• VWFpp/VWF:Ag and FVIII:C/VWF:Ag ratios define the pathophysiological mechanisms that play a role in VWD and various VWF mutations.

• A high VWFpp/VWF:Ag ratio indicates increased clearance of VWF and a high FVIII:C/VWF:Ag ratio decreased synthesis of VWF.

Abstract

During posttranslational modifications of von Willebrand factor (VWF), the VWF propeptide (VWFpp) is cleaved. The ratio between VWFpp and VWF antigen (VWF:Ag) and the ratio between factor VIII (FVIII:C) and VWF:Ag may be used to assess synthesis and clearance of VWF. We analyzed the contribution of VWFpp and ratios of VWFpp/VWF:Ag and FVIII:C/VWF:Ag in the pathophysiological characterization of type 1 von Willebrand disease (VWD) in the Molecular and Clinical Markers for the Diagnosis and Management of Type 1 VWD (MCMDM-1VWD) study. The VWFpp/VWF:Ag and FVIII:C/VWF:Ag ratios were increased among patients compared with unaffected family members and healthy controls. The VWFpp/VWF:Ag ratio was higher in individuals heterozygous for missense mutations than in those heterozygous for null alleles. In contrast, the FVIII:C/VWF:Ag ratio was highest among heterozygotes for VWF null alleles. The ratios of VWFpp/VWF:Ag and FVIII:C/VWF:Ag indicate that the pathophysiological mechanisms of type 1 VWD include reduced production and accelerated clearance of VWF, but that often a combination of both mechanisms is implicated.

Introduction

von Willebrand disease (VWD) is a bleeding disorder caused by inherited quantitative (types 1 and 3) or qualitative (type 2) defects of von Willebrand factor (VWF).¹ VWF supports platelet adhesion and carries factor VIII (FVIII).² VWF is synthesized as a pre-pro-VWF precursor protein. After cleavage of the signal peptide, the pro-VWF dimerizes in the endoplasmic reticulum (ER) via disulfide bridges at the carboxy-terminal cysteine knot (CK).² In the Golgi, pro-VWF dimers multimerize via disulfide bridges between D′D3 domains. Subsequently, the VWF propeptide (VWFpp) is cleaved but stays noncovalently attached to VWF, only to become dissociated after release into the circulation.² VWF and VWFpp are secreted in equimolar amounts.^2,3

Because of the different half-lives of VWFpp (2 hours) and VWF (8-12 hours), the ratio between VWFpp and VWF antigen (VWF:Ag) in plasma can be used to assess synthesis, secretion, and clearance rates of VWF.^4-7 We have reported an increased VWFpp/VWF:Ag ratio in 19 individuals of the Molecular and Clinical Markers for the Diagnosis and Management of Type 1 VWD (MCMDM-1VWD) study in whom the response to desmopressin showed a decreased VWF half-life.⁸ Some other type 1 VWD patients with increased clearance of VWF could be identified by an increased VWFpp/VWF:Ag ratio.^9,10

While VWFpp and VWF are cleared independently, FVIII is in complex with VWF and their half-lives are related. The ratio between FVIII coagulant activity (FVIII:C) and VWF:Ag (FVIII:C/VWF:Ag) may be used in addition to the VWFpp/VWF:Ag ratio—FVIII:C/VWF:Ag is increased when VWF synthesis is reduced but the ratio remains 1 when VWF is cleared faster.^11-13 Opposing results are obtained for VWFpp/VWF:Ag, which will remain unchanged with reduced synthesis, but will increase with reduced half-life of VWF.

Using the VWFpp/VWF:Ag and FVIII:C/VWF:Ag ratios, we have investigated the pathophysiological mechanisms involved in type 1 VWD.

Methods

In the European MCMDM-1VWD study, 744 individuals, including index cases (ICs), affected family members (AFMs), and unaffected family members (UFMs), from 154 families previously diagnosed with type 1 VWD as well as healthy controls (HCs) were recruited. Local ethical review committees approved the study. Written informed consent was obtained in accordance with the Declaration of Helsinki. The evaluation of phenotype and genotype was reported previously.^8,14-20 VWF:Ag, VWF ristocetin cofactor activity (VWF:RCo), FVIII:C, VWF multimers, ABO blood group genotypes, and VWF mutations were determined as described.^14-16,19,21 VWFpp antigen was measured in 387 HCs and in all remaining ICs (n = 137), AFMs (n = 264), and UFMs (n = 290) for whom plasma samples were available by enzyme-linked immunosorbent assay using antibodies from Sanquin (Amsterdam, The Netherlands).⁴

Statistical analyses were performed using SPSS Statistics 20.0 (IBM Corporation, Armonk, NY). As phenotypic data were not normally distributed, even after logarithmic transformation, median values and interquartile ranges were used as indices of centrality and dispersion. Differences were tested by nonparametric tests (Mann-Whitney U test, Kruskal-Wallis test). Results were plotted using GraphPad Prism 4.00 (GraphPad Software, San Diego, CA).

Results and discussion

It has been described in a selected subset of VWD patients that the VWFpp/VWF:Ag ratio can identify VWF with decreased half-life.^8-10 We now report on VWFpp and the ratios of VWFpp/VWF:Ag and FVIII:C/VWF:Ag in the entire MCMDM-1VWD study (supplemental Table 1, available on the Blood website). VWFpp was lower in ICs and AFMs than in HCs (P < .001), although less marked than the VWF:Ag reduction. VWFpp/VWF:Ag was increased for ICs and AFMs compared with HCs and this was also observed for FVIII:C/VWF:Ag (P < .001).

ABO blood group determines VWF and FVIII levels via VWF clearance.^22,23 As a reflection of increased clearance FVIII:C, VWF:Ag and VWF:RCo levels were lower (P < .001) in HCs with blood group O compared with non-O (supplemental Table 2). Blood group did not influence FVIII:C/VWF:Ag, supporting the view that VWF and FVIII are cleared as a complex.¹³ No difference was observed in the VWF:RCo/VWF:Ag ratio between HCs with blood group O and non-O, indicating a proportional decrease of VWF:Ag and VWF:RCo in blood group O and no suggestion of preferential clearance or proteolysis of large multimers in HCs with blood group O. In HCs, the VWFpp levels were not influenced by blood group, whereas VWFpp/VWF:Ag was clearly increased for blood group O in line with more rapid clearance of mature VWF.⁷

VWFpp and ratios of VWFpp/VWF:Ag, FVIII:C/VWF:Ag and VWF:RCo/VWF:Ag were analyzed after stratification for multimer pattern and mutation status (Table 1, Figure 1). In the MCMDM-1VWD cohort, subtle multimer abnormalities were reported in 38% of the ICs, and these individuals had more severe phenotypes, higher penetrance, a greater extent of linkage to the VWF gene locus, and mutations identified in all cases.^14,16,19,21 The increased VWFpp/VWF:Ag ratio was particularly raised (median 4.3) in patients with abnormal multimers and mutations (Table 1). An increased VWFpp/VWF:Ag ratio was a good predictor of VWD patients with mutations in the VWF gene (Figure 1A): a VWFpp/VWF:Ag >3 had a positive predictive value for the presence of a VWF mutation of 98% with a specificity of 99% in the entire cohort of patients and family members. The VWF:RCo/VWF:Ag ratio was decreased in the group with abnormal multimers (median 0.6, P < .001); however, the large coefficient of variation for VWF:RCo assays reduces the significance of this ratio when VWF levels are low. Among the UFMs, there were 31 individuals classified as unaffected but with a VWF mutation identified (Table 1). Those nonpenetrant cases had slightly higher VWFpp/VWF:Ag and FVIII:C/VWF:Ag ratios than UFMs without a mutation (P values .013 and .008 respectively, Table 1). Patients with mutations in families showing complete cosegregation between VWF gene and disease phenotype had higher VWFpp/VWF:Ag and FVIII:C/VWF:Ag ratios than patients from families with incomplete cosegregation (Table 1).

Table 1.

Stratification by multimer pattern and mutation status, linkage, and type of mutation

	Multimer pattern19	Mutation identified	n	VWFpp,U/dL	VWFpp/VWF:Ag	FVIII:C/VWF:Ag	VWF:RCo/VWF:Ag
IC + AFM	Normal	No	113	92 (81-108)	1.6 (1.3-1.9)	1.4 (1.1-1.7)	1.1 (0.9-1.2)
		Yes	134	81 (64-101)	2.0 (1.5-2.4)	1.7 (1.2-2.3)*	1.0 (0.9-1.2)
	Abnormal	Yes	150	79 (65-100)	4.3 (2.9-7.0)	1.8 (1.2-2.5)	0.6 (0.3-0.9)†
UFM	Normal	No	255	111 (98-128)	1.2 (1.0-1.5)	1.2 (0.9-1.4)	1.0 (0.8-1.2)
		Yes	31	115 (97-132)	1.4 (1.1-1.8)	1.3 (1.1-1.7)	0.9 (0.8-1.2)
Linkage of mutation‡
Cosegregation			174	78 (65-99)	3.3 (2.0-5.4)	1.9 (1.3-2.6)§	0.8 (0.4-1.1)
No cosegregation			57	80 (63-103)	2.4 (1.6-4.0)	1.6 (1.0-2.0)	1.0 (0.6-1.2)║
P cosegregation vs no cosegregation				0.786	0.048	0.003	0.11
Type of mutation
Missense mutations			228	89 (73-109)	2.8 (1.8-5.3)	1.6 (1.1-2.1)¶	0.8 (0.5-1.1)¶
“Null” mutations#			20	72 (63-85)	2.0 (1.7-2.4)	2.4 (2.0-2.8)	1.1 (1.0-1.3)
HC			387	118 (104-136)	1.2 (1.0-1.5)	1.1 (1.0-1.3)**	1.1 (0.9-1.3)
P missense vs null				0.001	0.007	<0.001	0.001

Results are indicated as median (25th to 75th percentile).

^*n = 133.

^†n = 149.

^‡Linkage was defined as complete cosegregation (pedigrees with no phenocopies and fully penetrant) or incomplete cosegregation (pedigrees with either phenocopies or nonpenetrance).¹⁴

^§n = 173.

^║n = 56.

^¶n = 227.

^#Comprise premature stop codons caused by nonsense mutations, frame shifts (small deletions and insertions), and out-of-frame splice site mutations.

^**n = 385.

Figure 1.

Ratios of VWF, VWFpp, and FVIII:C. (A) Ratios of VWFpp/VWF:Ag, FVIII:C/VWF:Ag, and VWF:RCo/VWF:Ag are shown for HCs, patients without a mutation identified (No mutation), and patients with a mutation identified in the VWF gene (Mutation). Patients represent the combined results of ICs and affected family members (IC + AFM); the ICs and AFMs were combined as there were no significant differences between the groups. The horizontal gray lines indicate median ratios. The dashed lines indicate the upper limit of the normal range for VWFpp/VWF:Ag (97.5th percentile is 2.2) and FVIII:C/VWF:Ag (97.5th percentile is 1.9) and the lower limit of the normal range for VWF:RCo/VWF:Ag (2.5th percentile is 0.6). For a better graphic representation, 4 outliers with very high ratios in the Mutation group for VWFpp/VWF:Ag and 2 in the Mutation group for FVIII:C/VWF:Ag were omitted. All groups differed significantly from each other (all comparisons P < .001 with the exception of VWF:RCo/VWF:Ag HC vs No mutation, P = .0224). (B) Scatter plots of FVIII:C/VWF:Ag vs VWFpp/VWF:Ag for missense mutations (n = 224, 4 outliers were omitted for better graphic representation), null mutations (n = 20), other mutations (n = 42), and for patients with no mutation (n = 115). The group of “other mutations” comprises putative splice site mutations and changes in the 5′ untranslated region that have not yet been proven by molecular studies to result in null alleles; however, the FVIII:C/VWF:Ag suggests indeed a defect of synthesis in many of them. (C) For all individuals heterozygous for a single missense mutation, the mean (+SD) VWFpp/VWF:Ag ratio is shown. At a few codons, different substitutions were identified (p.C1130F/R/G, p.R1374H/C, p.C2477S/Y) that are listed separately as the VWFpp/VWF:Ag ratio differed for each amino acid substitution. The number of individuals carrying a specific mutation ranged from 1 to 27.

VWFpp in heterozygotes for VWF null mutations was approximately half of that in HCs, reflecting reduced synthesis (Table 1). A high FVIII:C/VWF:Ag ratio (median 2.4), identifying reduced VWF synthesis, was seen in carriers of null alleles (Table 1, Figure 1B). The FVIII:C/VWF:Ag ratio was above the upper limit of the normal range (1.9) in 80% of these heterozygotes. Missense mutations demonstrated the highest VWFpp/VWF:Ag ratio, reflecting faster clearance of abnormal mutant VWF, although not all missense mutations lead to increased clearance (Table 1, Figure 1B). In heterozygotes for VWF missense mutations, there was an intermediate reduction in VWFpp level and increase in FVIII:C/VWF:Ag ratio (median 1.6) indicating that missense mutations partly also cause reduced synthesis and/or secretion. Thus, missense mutations can cause a combined defect of reduced synthesis and increased clearance (Figure 1B). The extent of increased clearance was not the same for all missense mutations, and the highest VWFpp/VWF:Ag ratios clustered in the VWF D3 and A1 domains (Figure 1C). Overall, the ratios in most patients with no mutation fell within the normal range (Figure 1A-B) and no clear mechanism could be deduced, but in some of those patients the defect was mainly of reduced synthesis whereas 1 patient had primarily increased clearance (Figure 1B). For a group of other mutations, including putative splice site mutations and changes in the 5′ untranslated region where the pathogenic mechanism had not been fully characterized at the molecular level, a defect in synthesis could be deduced in the majority (Figure 1B).

In conclusion, the VWFpp/VWF:Ag and FVIII:C/VWF:Ag ratios define the pathophysiological mechanisms that play a role in VWD and various types of VWF mutations. Clinical implications of these findings range from identifying increased clearance of VWF, which is important when considering desmopressin treatment, to discrimination between low VWF levels due to blood group O versus VWF null alleles.

Appendix: study group members

The members of the MCMDM-1VWD Study Group are:

Ian Peake, Sheffield, UK; Anne Goodeve, Sheffield, UK; Francesco Rodeghiero, Vicenza, Italy; Giancarlo Castaman, Vicenza, Italy; Alberto Tosetto, Vicenza, Italy; Augusto B. Federici, Milano, Italy; Javier Batlle, La Coruna, Spain; Dominique Meyer, Paris, France; Edith Fressinaud, Nantes, France; Claudine Mazurier, Lille, France; Jenny Goudemand, Lille, France; Jeroen Eikenboom, Leiden, The Netherlands; Reinhard Schneppenheim, Hamburg, Germany; Ulrich Budde, Hamburg, Germany; Jørgen Ingerslev, Aarhus, Denmark; Zdena Vorlova, Prague, Czech Republic; David Habart, Prague, Czech Republic; Lars Holmberg, Lund, Sweden; Stefan Lethagen, Malmö, Sweden; John Pasi, Leicester, UK; and Frank Hill, Birmingham, UK.

Authorship

Contribution: J.E. designed the study, collected data, performed laboratory analyses, analyzed and interpreted results, was lead author of the initial manuscript, revised manuscript drafts, and reviewed and approved the final manuscript; A.B.F., R.J.D., U.B., R.S., J.B., M.T.C., and J.G. designed the study, collected data, performed laboratory analyses, and reviewed and approved the final manuscript; G.C. designed the study, collected data, performed laboratory analyses, revised manuscript drafts, and reviewed and approved the final manuscript; F.R. initiated, coordinated, and designed the study, collected data, performed laboratory analyses, and reviewed and approved the final manuscript; I.P. initiated, coordinated, and designed the study, collected data, performed laboratory analyses, revised manuscript drafts, and reviewed and approved the final manuscript; and A.G. initiated, coordinated, and designed the study, collected data, performed laboratory analyses, analyzed and interpreted results, revised manuscript drafts, and reviewed and approved the final manuscript.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

The current affiliation for A.B.F. is Hematology and Transfusion Medicine, L. Sacco University Hospital and Department of Clinical and Community Sciences, University of Milan, Milan, Italy.

A list of the the MCMDM-1VWD Study Group members appears in the “Appendix.”

Correspondence: Jeroen Eikenboom, Einthoven Laboratory for Experimental Vascular Medicine, Department of Thrombosis and Hemostasis, C7-Q, Leiden University Medical Center, PO Box 9600, 2300 RC Leiden, The Netherlands; e-mail: h.c.j.eikenboom@LUMC.nl.