Journal Club – vWD

Atiq, F., Blok, R., van Kwawegen, C. B., Doherty, D., Lavin, M., van der Bom, J. G., … & O’Donnell, J. S. (2024). Type 1 VWD classification revisited: novel insights from combined analysis of the LoVIC and WiN studies. Blood, 143(14), 1414-1424.

Clinical Question

Does low von Willebrand factor (VWF) constitute a discrete clinical entity distinct from Type 1 von willebrand disease (VWD) or does it represent a continuum within type 1 VWD, including age-related normalization of VWF levels?

Background

Type 1 VWD is a distinct subtype of VWD characterised by partial quantitative reduction in plasma VWF antigen (VWF:Ag). Prior guidelines made the distinction between type 1 VWD defined by VWF:Ag levels <30 IU/dL and an entity called low VWF:Ag, which was defined as plasma VWF:Ag between 30 to 50 IU/dL. This distinction was born out of prior studies which suggested that there were specific differences between these two patient cohorts. Specific examples included the fact that pathogenic VWF variants were more commonly identified in individuals with VWF:Ag <30 IU/dL and the observation that those with levels 30-50 IU/dL had a more variable bleeding phenotype and fewer genetic variants.

This distinction was, however, challenged in guidelines published by the American Society of Hematology/International Society on Thrombosis and Hemostasis/World Federation of Hemophilia/National Haemophilia Foundation in 2021. They recommended that VWF levels <30 IU/dL were sufficient to establish a diagnosis of type 1 VWD regardless of bleeding history provided qualitative defects are excluded (as evidenced by VWF activity/VWF:Ag >0.7). They also suggested that individuals with a significant bleeding history and plasma VWF plasma levels between 30-50 IU/dL also be diagnosed with type 1 VWD. They also de-emphasized the use of the term “low VWF”. This recommendation was made with the intention of avoiding under diagnosis and to ensure appropriate access to care, particularly for patients with bleeding symptoms.

This decision has been challenged since that time and has been the object of several studies. It has been further complicated by the observation that VWF:Ag levels increase with aging, the mechanism of which is not completely understood but proposed to be related to increased biosynthesis and decreased clearance of VWF.

This article reports the combined analysis of the type 1 VWD and low VWF populations in the Low VWF in Ireland Cohort (LoVIC) and Willebrand in the Netherlands (WiN) cohort studies. It was aimed at exploring whether low VWF constituted a distinct clinical entity or if it represented an evolving VWD phenotype that occurs with aging.

Study Design

• The LoVIC study was a national prospective cohort study initiated in 2014 and performed in Ireland. The inclusion criteria included historically lowest VWF:Ag and or VWF:Act in the 30-50 IU/dL range typically confirmed on repeat testing. Pregnant patients were excluded from the study.
• The WiN study was also a national multicenter observational cohort study. It was performed in the Netherlands and initiated in 2007. Inclusion criteria included the presence of a personal or family history of bleeding with reduced VWF levels
• generally <30 IU/dL. Patients who had other bleeding disorders present were excluded.
• Bleeding phenotypes were assessed using validated bleeding assessment tools, including the Condensed Molecular and Clinical Markers for the Diagnosis and Management of Type 1 VWD (Condensed MCMDM-1VWD) score, which was utilised in both cohorts and the ISTH-BAT used in the LoVIC study.
• Additional data collected included baseline laboratory hemostasis studies including plasma VWF:Ag, VWF ristocetin cofactor, VWF:CB, and FVIII activity. VWF propeptide levels were also measured in a subset of patients.
• A subset of patients in both studies were also followed over time with serial VWF and FVIII levels with those followed for >5 years being a part of the cohort evaluated for the effect of aging on plasma VWF levels.
• Desmopressin trials were also performed on a subset of patients with a dose of 0.3 mcg/kg body weight administered intravenously or 300 mcg administered intranasally, with blood levels taken at predetermined time points and responses assessed using standard criteria.
• Genetic analysis was also done on the cohorts of both studies including sequencing of the VWF gene to assess for pathogenic variants

Populations

• Baseline characteristics
  • A total of 565 patients were included: 403 from the WiN cohort and 162 patients with low VWF from the LoVIC cohort.
  • o The WiN patients were further categorized based on longitudinal VWF levels into three groups:
    • Persistent VWF levels <30 IU/dL (n=188)
    • Partial correction of plasma VWF levels into the 30-50 IU/dL range (n=121)
    • Normalization of plasma VWF levels >50 IU/dL (n=94)
• General demographic observations
  • The LoVIC cohort was predominantly female (88.9%) whereas the WiN subgroups included a lower proportion of females (63.3%, 69.4% and 69.1% in the persistent <30 IU/dL, 30-50 IU/dL and normalized >50 IU/dL groups respectively)
  • The LoVIC cohort included more blood group O patients (91.8%) compared to the WiN subgroups where blood group O made up 61.8%, 76% and 73.1 % of persistent <30 IU/dL, 30-50 IU/dL and normalized >50 IU/dL groups respectively.
  • No statistically significant differences were noted in VWF parameters were observed by sex or ABO blood group.
  • The MCMDM-1 VWD bleeding score and as well as reports of bleeding requiring treatment in the year leading up to study participation in the WiN study were not different in males and females across the WiN subgroups.
• Specific age-related observations
  • Mean age at diagnosis in the LoVIC cohort was significantly higher than the 3 WiN subgroups (WiN persistent VWF<30 IU/dL, WiN partially corrected 30-50 IU/dL and WiN normalized >50 IU/dL) at 32.5 yrs vs 23.3 yrs vs 26.5 years and 25.1 years respectively. There was also no significant difference in the age at enrollment between the cohorts.
  • An age-dependent increase in plasma VWF levels was observed in a substantial proportion of patients with type 1 VWD (the WiN patients) and in the majority of those with low VWF (the LoVIC patients).
  • In the WiN cohort, 47% of patients had VWF levels remaining <30 IU/dL despite advancing age.
  • 30% of these patients, however,r had an increase in plasma VWF levels into the low VWF range over time, and 23% had age dependent increments that led to complete normalization of VWF levels to >50 IU/dL.
  • In the LoVIC cohort, the plasma VWF levels normalized to >50 IU/dL over time in 39% of patients.
  • It was also observed that there was substantial overlap in the VWF levels between patients with low VWF and those with type 1 VWD whose plasma levels increased to >30 IU/dL as they aged.
  • On subcategorisation of the WiN cohort into the 3 groups it appeared that the LoVIC cohort showed similar characteristics to the WiN normalized >50 IU/dL subgroup. It was also seen that the increase in the VWF:Ag per year was at a similar rate in the LoVIC and the WiN normalized >50 IU/dL patient groups. Multiple regression analysis was also done and suggested that differences in VWF levels between the LoVIC and WiN cohorts were largely attenuated after adjusting for age at diagnosis.
  • When taken together, the results indicate significant heterogeneity among patients with type 1 VWD with respect to the effect of aging on their plasma VWF:Ag levels. The results suggest that it may have been possible that some patients with low VWF may have had lower VWF levels earlier in life that would have met criteria for type 1 VWD.
• Further pathophysiologic observations
  • Within the LoVIC population and WiN subgroups it was found that FVIII:C/VWF:Ag ratios were similar in the WiN normalized >50 IU/dL and LoVIC patient groups
  • WiN patients with persistent VWF<30 IU/dL had higher FVIII:C/VWFAg ratios than WiN normalized and LoVIC patients
  • A similar observation was noted with VWF propertied to VWF:Ag ratios, with the ratios being similar in the WiN normalized and LoVIC patient groups but markedly elevated in the WiN patients with persistent VWF<30 IU/dL compared to the other groups.
  • These findings suggest that the pathophysiological mechanisms may be similar in the WiN normalized and LoVIC groups and that patients with persistently low VWF (<30 IU/dL) may have a pathophysiological profile related to increased VWF clearance.

Intervention and Outcomes

• Response to Desmopressin
  • A complete response to desmopressin was observed in 100% of the patients in the LoVIC group as well as those in the WiN patients whose VWF had increased >30 IU/dL. Fewer patients in the WiN group with persistent VWF<30 IU/dL demonstrated a complete response (58.1%). This subset of patients also demonstrated a significantly attenuated response at 1hr after and a more rapid decline in VWF levels over time.
  • VWF:Ag plasma levels at 1, 4 and 24 hours after desmopressin were higher in the WiN normalized cohort compared to the LoVIC patients. The authors hypothesized that this may have been related to older age of the WiN- normalized cohort. To further explore this, they assessed desmopressin responses across age groups and found that the responses at 1, 4 and 24 hours increased with advancing age.
  • These findings suggest that patients with low VWF and normalized type 1 VWD share similar pathogenesis and that they exhibit a comparable response to desmopressin. They also suggest that age may influence the magnitude and persistence of the VWF response following desmopressin administration.

• Determinants of age-induced VWF level normalization in low VWF and type 1 VWD
  • Longitudinal analysis was performed in the LoVIC and WiN patients who had been followed for >5 yrs.
  • They found no difference in the VWF:Ag increase per decade between the 2 groups
  • Normalization in plasma VWF with aging was more common in patients with low VWF than in patients with type 1 VWD
  • 66% of patients in the LoVIC cohort achieved a VWF level >50 IU/dL over time
  • 51.9% of patients in the WiN type 1 VWD cohort achieved normalization despite longer follow up (16.9+/- 7.4 years)
  • None of the patients who had an incomplete response to desmopressin achieved plasma VWF>50 IU/dL during follow up, while the median time to normalization among responders was 13 years, suggesting that a poor desmopressin response may be a strong predictor of failure to normalize VWF levels with aging.

• Additional findings included:
  • Pathogenic VWF variants were more common in the WiN subgroup of patients with persistent VWF levels<30 IU/dL
  • There was no significant difference in the percentage of patients with the VWF gene variants between LoVIC and WiN partially corrected or WiN normalized >50 IU/dL subgroups.
  • The presence of a pathogenic VWF gene variant was significantly associated with a lower likelihood of age-related VWF normalization.
  • Historically lowest VWF levels <10 IU/dL was strongly associated with significantly reduced normalization of VWF levels over time

Commentary and Intepretation:

In combining and analysing data from the LoVIC and WiN cohorts, Atiq et al. presented evidence that type 1 VWD is a biologically heterogenous disorder. Their findings challenge the traditional view that low VWF and type 1 VWD are distinct entities but instead suggests that these categories lie on a continuum of quantitative VWF deficiency.

The data suggests that individuals with low VWF share clinical and laboratory features with a subset of patients with type 1 VWD, particularly those whose levels increase with age, raising the possibility that low VWF may represent an evolving or age-dependent phenotype within the spectrum of type 1 VWD.

Conversely, the data also suggest that patients with type 1 VWD with persistent VWF levels <30 IU/dL despite aging may be a biologically distinct subgroup characterized by differences in VWF kinetics and responsiveness to desmopressin.

The study also highlights potential predictors of longitudinal VWF trajectories, specifically the fact that reduced responsiveness to desmopressin and the presence of pathogenic VWF variants were associated with a lower likelihood of VWF normalization over time.

There are several limitations to the study that must be considered when interpreting the findings. The fact that two cohorts with different inclusion criteria were combined and analysed introduces the potential for selection bias and does limit the direct comparability between the groups. The overrepresentation of female patients and those with blood group O in the LoVIC cohort could also complicate interpretation of the demographic and phenotypic differences. In addition, bleeding outcomes were not the primary focus of the analysis, and it remains unclear whether biochemical normalization of VWF with age translates to meaningful changes in bleeding risk.

In conclusion, these findings support a shift away from the previously-held view of type 1 VWD as being a distinct clinical entity to low VWF and suggests moving towards a more biology-informed view of type 1 VWD as a heterogenous and dynamic disorder in which longitudinal phenotype and responsiveness to desmopressin may be as important as single baseline VWF measurements.

Jason Lofters MBBS completed medical school at the University of the West Indies, Mona (Jamaica) and then internal medicine residency at Englewood Hospital/Hackensack University Medical Center in New Jersey, where he also served as chief resident. He is a current chief fellow at Boston University Medical Center. His clinical and research interests focus on classical hematology with particular emphasis on red blood cell disorders.