Type 3 von Willebrand Disease (VWD) is the rarest and most severe form of VWD and is characterized by the complete absence of von Willebrand factor (VWF). Inheritance is autosomal recessive although heterozygous carriers may be affected with type 1 VWD. As part of the Zimmerman Program for the Molecular and Clinical Biology of VWD (Zimmerman Program) we sought to explore the molecular pathogenesis, correlate bleeding phenotype and severity, and determine the presence and impact of alloantibodies in type 3 VWD.

57 index cases (IC) with a pre-existing diagnosis of type 3 VWD by their local treatment center were enrolled into the Zimmerman Program. Clinical VWF testing for phenotypic analysis was done in a central laboratory at BloodCenter of Wisconsin and included FVIII, VWF:Ag, VWF:RCo, VWFpp, VWF:CB and multimer analysis. Full-length exonic VWF sequencing was performed, including intron/exon boundaries. If two sequence variants (SV) were not identified, cases were further investigated by array comparative genomic hybridization (aCGH) to identify any large deletions or duplications in the VWF gene. Bleeding symptoms were quantified using the ISTH bleeding score (ISTH BS). The presence of VWF alloantibodies was determined in type 3 index cases and affected family members (AFM) using an ELISA based assay that detected IgM and IgG antibodies.

Of the 57 IC with type 3 VWD enrolled, 42 (74%) had central laboratory testing confirming type 3 (absence of VWF), 7 subjects (12%) had detectable levels and were re-classified as type 1-severe, and 8 subjects (14%) had a significantly increased VWFpp/VWF:Ag and were re-classified as type 1C. Sequencing results were available on 40 of the confirmed 42 type 3 IC. SV were identified in 37 of these subjects (93%): 31 subjects had 2 SVs, 6 subjects had only 1, and 3 had none; 68 of the 80 SVs expected were identified (85%). Of the SVs identified, 37% were frameshift, 29% nonsense, 15% missense, 9% splice site, 7% large deletions detected only by aCGH, and 3% intronic. The large deletions included heterozygous deletions of exon 1-3, exon 4-5, exon 18 and exon 35-38 in combination with another SV. One case only had 2 different deletions (exon 1-3 and exon 4-5) on separate alleles.

An equal number (21) of females and males comprised the type 3 cohort with a mean age at enrollment of 24 (range 0-61). Median bleeding score in this group was 15, with no difference between males and females (15). Adults (n=25) had a median ISTH BS of 18 that was significantly increased (p<0.0005) compared to a score of 11 in pediatric subjects (<18 years old, n=17). No difference in ISTH BS was observed according to number of SV or between those with or without null alleles. The highest median bleeding subscores reported were hemarthrosis (3), followed by nose bleeds (2), bruising (2), and oral cavity bleeding (2) and bleeding from minor wounds (1).

51 type 3 subjects, including 42 IC and 9 AFM, were tested for alloantibodies to VWF. 16 subjects showed positivity on two separate assays to either IgM or IgG using a 3 SD cutoff above the mean of normal pooled plasma: 4 were markedly positive and 12 were slightly positive. Of the 4 markedly positive subjects, 1 subject had IgM antibodies and 3 had IgG. 3 of the subjects with alloantibodies had no SVs (acquired VWD ruled out), while 1 subject had 2 nonsense variants. ISTH BS was not different between the 16 subjects with an alloantibody (16.5) compared to those without (16). Longitudinal studies will determine the correlation between treatment with VWF concentrate and development of alloantibodies in type 3 VWD.

In this type 3 VWD cohort, we have found that some type 1-severe and type 1C patients have been misdiagnosed as type 3. We have also observed that not all type 3 subjects have 2 SV and in fact, 15% of type 3 single alleles did not have an identifiable SV. Null alleles including nonsense, frameshift and large deletions account for 73% of the variant alleles in this cohort. Bleeding symptoms are severe in type 3 and are predominantly due to hemarthrosis, nose bleeds, bruising, oral cavity bleeding and minor wound bleeding. While alloantibodies have been previously associated mainly with homozygous large deletions, our data suggest that type 3 VWD, regardless of the genetic cause, carries a risk of alloantibody formation.


Flood: CSL Behring: Consultancy; Shire: Consultancy. Gill: Shire: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Bayer: Consultancy, Membership on an entity's Board of Directors or advisory committees; CSL Behring: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding.

Author notes


Asterisk with author names denotes non-ASH members.