Type 1 von Willebrand Disease in a Pediatric Patient Caused by a Novel Heterozygous Deletion of Exons 1 to 6 of the von Willebrand Factor Gene: A Case Report

Abstract

A 6-year-old boy was referred to a hematologist due to excessive mucocutaneous bleeding. Diagnostic assessment for von Willebrand disease (VWD) was indicated and included both coagulation and genetic testing. Laboratory testing revealed proportionally decreased von Willebrand factor (VWF) glycoprotein Ib-binding activity (23.6%) compared to VWF antigen (24.7%), similarly decreased VWF collagen-binding activity (24.2%), and normally distributed VWF multimers, with decreased intensity of all fractions. Diagnosis of type 1 VWD was established. Genetic analysis by means of next-generation sequencing (NGS) of VWF and coagulation factor VIII genes did not identify any causative mutations. Additionally, multiplex ligation-dependent probe amplification (MLPA) of VWF gene exons revealed a heterozygous deletion of exons 1 to 6, which is reported in type 1 VWD for the first time. Application of MLPA was crucial for revealing the genetic basis of type 1 VWD in this case, which would have remained undetected if only NGS was used.

von Willebrand disease, bleeding, von Willebrand factor, coagulation testing, next-generation sequencing, multiplex ligation-dependent probe amplification

von Willebrand disease (VWD) is the most common inherited bleeding disorder caused by quantitative or qualitative changes of von Willebrand factor (VWF), a large multimeric glycoprotein with 2 crucial roles in hemostasis. It mediates platelet adhesion and aggregation at the site of vascular injury through interaction with the platelet receptor glycoprotein Ib (GPIb), as well as transports and stabilizes coagulation factor VIII (FVIII) in circulation.¹ The gene encoding VWF is located on the short arm of chromosome 12, spans 178 kilobases of genomic sequence, and consists of 52 exons.²

Type 1 is the most frequent form of VWD, affecting up to 80% of all persons with VWD, and is characterized by variable degrees of quantitative deficiencies involving structurally and functionally normal VWF.^1,3,4 The majority of patients present with mild mucocutaneous bleeding symptoms, including epistaxis, easy bruising, and prolonged bleeding following minor injuries, whereas severe bleeding symptoms are rare and related to cases with significantly reduced VWF levels.^4,5 Key laboratory findings in type 1 VWD comprise proportionally reduced VWF activity and antigen levels (VWF:Ag), resulting in a ratio of the 2 above 0.7. The VWF multimers are normally distributed, with equally decreased intensity of all fractions.¹

Type 1 VWD shows an autosomal dominant inheritance pattern, and its genetic basis is largely heterogeneous. To date, around 140 different mutations associated with type 1 VWD have been identified throughout the whole VWF gene.⁶ Missense mutations are found to be the most common genetic cause, accounting for about 70% of cases, whereas splice, transcriptional, nonsense, and frameshift mutations constitute the remainder.⁷ Reduced levels of VWF in type 1 VWD are a consequence of altered molecular processing mechanisms that either result in increased clearance or decreased cellular secretion of VWF.⁷ However, phenotypic presentation of type 1 VWD can be largely variable, even among members of the same family or unrelated patients with the same mutations, due to their incomplete penetrance and variable expressivity.⁸ Genetic characterization of patients with type 1 VWD requires analysis of the entire VWF gene coding region, which became possible with the introduction of next-generation sequencing (NGS) capable of reliably detecting point and small frameshift mutations.⁹ In addition, a minority of type 1 VWD cases are reportedly caused by large heterozygous deletions and duplications. To correctly detect and identify those cases, gene dosage analysis, particularly multiplex ligation-dependent probe amplification (MLPA), should be used.^2,9,10

Here, we present a case of a pediatric patient with bleeding symptoms and coagulation testing results indicative of type 1 VWD, in whom MLPA analysis was crucial for identification of the underlying genetic cause.

Materials and Methods

Patient Presentation and Blood Sampling

A 6-year-old boy without a known family history of bleeding disorders was referred to a hematologist at the Department of Pediatrics Hematology and Oncology, University Hospital Center Zagreb, Zagreb, Croatia, due to recurrent episodes of epistaxis, prolonged bleeding following trivial injuries, and excessive bruising after minor trauma. Laboratory testing for VWD was indicated and included both coagulation and genetic analyses. For this purpose, two 2.7 mL 3.2% trisodium citrate and one 2 mL EDTA vacutainers (Becton Dickinson) were drawn. The study was approved by the University Hospital Center Zagreb Ethics Committee (8.1-19/293-2, 02/21 AG), and the patient’s mother gave written informed consent for anonymous publication of medical data.

Coagulation Testing

Capacity of primary hemostasis was determined from whole blood samples, within 2 hours from blood draw, on the platelet function analyzer-200 (PFA-200) using dedicated cartridges coated with platelet agonists collagen and both epinephrine and adenosine-diphosphate.

For other coagulation assays, platelet-poor plasma was obtained by double centrifugation for 15 minutes at 2000g. Prothrombin time was determined using the recombinant thromboplastin Dade Innovin and activated partial thromboplastin time with Dade Actin FS and calcium chloride on the Sysmex CS-5100 coagulation analyzer. The VWF activity was measured on an Atellica COAG 360 analyzer by means of the automated immunoturbidimetric assay INNOVANCE VWF Ac that uses microparticles coated with GPIb containing 2 gain-of-function mutations that enhance VWF binding (VWF:GPIbM), thus reflecting the GPIb-binding activity of VWF. The VWF:Ag was determined using the automated immunoturbidimetric assay VWF Ag, and FVIII activity was measured by a 1-stage clotting assay, using FVIII-deficient plasma and Dade Actin FS, on the Atellica COAG 360 coagulation analyzer. All respective reagents and analyzers are produced by Siemens Healthcare, and all analyses were performed following the original manufacturer’s protocols.

The VWF collagen-binding activity (VWF:CB) was measured using the enzyme-linked immunosorbent assay Technozym vWF:CBA (Technoclone) that reflects the ability of VWF to bind human collagen type III. The ratios VWF:GPIbM/VWF:Ag and VWF:CB/VWF:Ag were calculated.

Multimeric analysis was performed by agarose gel electrophoresis with direct immunofixation using the Hydragel 5 von Willebrand multimers assay kit on the Hydrasys 2 Scan instrument (Sebia). The results were interpreted in relation to the multimeric profile of normal plasma with values of VWF:Ag ~ 100%, as recommended by the manufacturer.

Genetic Analyses

Genomic DNA was isolated from EDTA whole blood using the automated magnetic bead-based method on the MagNA Pure Compact instrument (Roche Diagnostics).

Genetic analysis included NGS of all 52 exons, intronic flanking regions, and the promoter of the VWF gene as well as all 26 exons and the promoter of the FVIII gene. The DNA samples were prepared for sequencing using the Illumina DNA Prep with Enrichment kit (Illumina), strictly following the manufacturer’s protocol.¹¹ The NGS was performed on Miseq (Illumina) using the 300-cycle Miseq reagent kit v2 with 150 bp paired end reads. Sequencing raw data were uploaded in the form of FASTQ files in the BaseSpace Variant Interpreter software (Illumina), where alignment to the hg19 reference genome was carried out.

For detection of large duplications or deletions within the VWF gene, MLPA analysis using the SALSA MLPA reagent kit and probe sets dedicated for VWF gene analysis SALSA MLPA Probemix VWF P011 and P012 (MRC Holland) was performed according to the manufacturer’s protocol. Amplification products were identified on the ABI-3130xl Genetic Analyzer (Applied Biosystems), and DNA dosage was estimated using Coffalyser.Net software (MRC Holland). Interpretation of results was based on the evaluation of changes in probe signal intensity in relation to reference samples; specifically, a decrease or increase in probes signal intensity compared to reference samples indicates a deletion or duplication, respectively.

Results

Results of coagulation testing are presented in Table 1 and indicated partial quantitative deficiency of VWF. Multimeric analysis showed proportional reduction of all VWF multimer fractions compared to normal plasma, as shown in Figure 1. Based on these results, diagnosis of type 1 VWD was established.

Table 1. 

Results of Coagulation Testing

Assay	Result	Reference Interval
PT (ratio)	1.01	>.70
aPTT (s)	32.6	20.0–30.0
VWF:GPIbM (%)	23.6	50.0–187.0
VWF:Ag (%)	24.7	50.0–160.0
VWF:GPIbM/ VWF:Ag (ratio)	0.96	N/A
FVIII (%)	54.0	50.0–149.0
VWF:CB (%)	24.2	40.0–250.0

aPTT, activated partial thromboplastin time; FVIII, coagulation factor VIII activity; N/A, not applicable; PT, prothrombin time; VWF:Ag, von Willebrand factor antigen; VWF:CB, von Willebrand factor collagen-binding activity; VWF:GPIbM, von Willebrand factor gain-of-function mutant glycoprotein-Ib binding activity.

Figure 1.

Results of multimers analysis showing a normal multimer distribution with proportional decrease of all multimer fractions. The gray line represents the patient’s multimer distribution, while the black line represents normal pooled plasma sample. LMWM, low-molecular-weight multimers; IMWM, intermediate-molecular-weight multimers; HMWM, high-molecular-weight multimers.

For identification of the genetic cause of VWD in this patient, first NGS analysis was performed. No disease-associated genetic variants were identified either within the VWF or the FVIII gene. Additional analysis using MLPA revealed the presence of only 1 copy of exons 1 to 6 of the VWF gene, as shown in Figure 2A and 2B. A ratio of 0.5 between signals of the analyzed probes for the respective exons in the patient’s sample and reference samples was obtained, thus indicating a deletion in the heterozygous state. For comparison, Figure 2C and 2D show results for a control sample with 2 copies of all analyzed exons, with all ratios being ~1.

Figure 2.

Results of multiplex ligation-dependent probe amplification showing the presence of 1 copy of exons 2, 5, and 6 using SALSA Probemix P011 VWF (A) and exons 1, 3, and 4 with SALSA Probemix P012 VWF (B). A ratio of 0.5 between signals of the analyzed probes for the respective exons in the patient’s sample and reference samples was obtained, as indicated by squared frames. For comparison, results of analysis of a control sample with 2 copies of exons analyzed with SALSA Probemix P011 VWF (C) and with SALSA Probemix P012 VWF (D) are shown. The obtained ratio for all exons in the control sample is ~1.

Discussion

The presented case represents a rare occurrence of a large frameshift mutation associated with type 1 VWD phenotype in a pediatric patient. To the best of our knowledge, this is the first report of a type 1 VWD caused by the heterozygous deletion of exons 1 to 6 within the VWF gene.^2,6

The obtained coagulation testing results that included proportionally reduced VWF:GPIbM, VWF:Ag, and VWF:CB, with values below 30%, as well as normal multimeric distribution characterized by decreased intensity of all fractions, indicated quantitative deficiency of VWF and, together with mild mucocutaneous bleeding symptoms, fulfilled the diagnostic criteria for type 1 VWD.¹² Rather unexpectedly, the initial genetic analysis by means of NGS failed to identify the causative mutation. It is known that only about 65% of patients with type 1 VWD have an underlying mutation in the VWF gene, the incidence of which increases with decreased levels of VWF.¹³ In this case, the levels of VWF below 30% clearly indicated that the observed deficiency of VWF must be caused by a genetic alteration that NGS failed to detect due to inherent method limitations. Therefore, we speculated that a quantitative change of 1 or more exons could be the cause of type 1 VWD phenotype in this case. For this purpose, MLPA analysis was performed and indeed revealed that the patient is a heterozygous carrier of a deletion spanning exons 1 to 6 of the VWF gene. The NGS is a polymerase chain reaction–based technique that could not detect this genomic abnormality due to the presence of 1 copy of the affected gene region that was regularly amplified and sequenced, consequently masking the absence of the second allele.^9,14 In contrast, MLPA is a molecular technique based on amplification of probes hybridized to targeted gene regions. A decrease in respective exon copy numbers allowed reliable detection of the large heterozygous deletions involving exons 1 to 6.^9,14,15

Most studies dealing with MLPA have predominantly included patients with type 3 VWD, and it is presumed that the incidence of large deletions in type 1 VWD is likely to be underestimated due to limited studies in this group of patients.¹⁶ In patients with type 1 VWD, heterozygous deletions of exons 3, 4 to 5, 26 to 34, 32 to 34, and 33 to 34 have been identified so far.² Deletion of exons 4 to 5 has been extensively studied by Sutherland et al,¹⁰ who reported that homozygotes invariably present with unmeasurably low VWF levels and severe bleeding symptoms, thus usually ending up classified as type 3 VWD. In heterozygotes, mild-to-moderate clinical and laboratory phenotypes characteristic for type 1 VWD were recorded. There was similar patient presentation in our case, which could be expected since the heterozygous deletion of exons 1 to 6 also affects the initial part of the VWF gene responsible for encoding part of the D1 domain within the VWF propeptide, hence causing decreased secretion and defective multimerization of the mutant VWF and resulting in type 1 VWD phenotype.^10,17 However, when such heterozygous deletion is part of a compound heterozygous genotype, as in the case described by Yadegari et al¹⁴ where heterozygous deletion of exons 1 to 5 was combined with a missense mutation, the patient’s phenotype is compatible with type 3 VWD. In fact, up to 80% of patients with type 3 VWD have a null allele mutation, which is either autosomal recessive and homozygous, or compound heterozygous, thus resulting in undetectably low VWF levels and severe bleeding phenotype.⁸ A similarly severe phenotype can be found in the minority of patients with type 1 VWD caused either by fully penetrant dominant mutations or more rarely, a recessive genotype. However, severe type 1 VWD cases differ from type 3 VWD by the presence of low but still detectable VWF levels.¹⁸ Differentiation of severe type 1 VWD from type 3 VWD is important for choosing the appropriate treatment approach as patients with severe type 1 VWD might be responsive to desmopressin, whereas the only therapeutic option for patients with type 3 VWD are plasma-derived VWF/FVIII concentrates due to the lack of endogenous VWF.¹⁹ This clearly emphasizes the complexity of the genetic basis of VWD and the importance of the detection of the underlying genetic cause, which can be properly elucidated only through the use of adequate molecular techniques. The choice of the most suitable molecular diagnostic approach should be based on detailed evaluation of clinical symptoms and coagulation testing results as well as family history data, if available.

In conclusion, in this case, the addition of MLPA analysis to the molecular diagnostic algorithm of VWD was crucial for revealing the genetic basis of type 1 VWD, which would have remained undetected if only NGS was used. Therefore, in patients with clear clinical and laboratory evidence of type 1 VWD in whom causative mutations are not detected using sequencing methods, large heterozygous deletions of 1 or more exons should be considered as a possible cause of VWD and therefore MLPA should be applied for their detection. Proper genetic characterization of VWD might be especially important in pediatric patients such as this one for whom no positive family history of bleeding disorders is known. Such elaborate mutational analysis in the setting of type 1 VWD not only unequivocally confirms the diagnosis but also elucidates the underlying pathophysiological mechanism of the disorder, reveals recurrence risks in family members, and provides differential diagnosis from other mild bleeding disorders with overlapping clinical symptoms but different etiology,⁸ hence providing the basis for proper future therapeutic management as well as genetic counselling.

Conflict of Interest Disclosure

The authors have nothing to disclose.

Abbreviations

Abbreviations
 
• VWD

  von Willebrand disease

 
• VWF

  von Willebrand factor

 
• NGS

  next-generation sequencing

 
• MLPA

  multiplex ligation-dependent probe amplification

 
• GPIb

  glycoprotein Ib

 
• FVIII

  coagulation factor VIII

 
• VWF:Ag

  von Willebrand factor antigen

 
• VWF:GPIbM

  von Willebrand factor gain-of-function mutant glycoprotein-Ib binding activity

 
• VWF:CB

  von Willebrand factor collagen-binding activity