The Complete Type I VWD Cohort Of The Zimmerman Program For The Molecular and Clinical Biology Of VWD - Phenotypic Assignment, Mutation Frequency, and Bleeding Assessment

The Zimmerman Program for the Molecular and Clinical Biology of VWD (ZPMCBVWD) is an NIH Program Project for the study of VWD in the USA and collaboration with ongoing studies in Canada and the University of Sheffield. Study subjects were recruited from 9 Primary Clinical Centers and 21 Secondary Clinical Centers across the USA and had to have the prior diagnosis of VWD on an intention to treat basis. We recruited 651 index cases, 2017 family members, and 247 normal controls. Of the index cases 152 had type 2 or 3 VWD or other variants and not part of this report. The remaining 499 represents the now completed Type 1 VWD cohort.

All 2915 study subjects underwent extensive laboratory testing including VWF:Ag, VWF:RCo, VWF multimeric analysis, FVIII activity, VWFpp, and VWF:CB (collagen binding), blood typing, and VWF linkage analysis. A detailed quantitative bleeding score (BS) was also performed on all 2915 individuals. Index cases had full length VWF sequencing with mutation confirmation and exclusion in family members. When indicated VWF:F8B (FVIII binding), VWF:PB (platelet binding to GPIb), VWF:IbCo (spontaneous binding to modified GPIb in absence of ristocetin), and repeat VWF testing in a second laboratory by different methods. When possible, historical laboratory results obtained at the time of initial diagnosis were obtained and entered into our Investig-8 Database.

Based upon clinical laboratory studies and phenotypic assignment, this cohort was found to include 232 type 1 VWD (VWF:Ag or VWF:RCo <40 IU/dL including 66 type 1C or severe type 1 VWD); 93 low-VWF (LVWF with lab studies between 40 IU/dL and the lower end of the normal range); 119 type 1H (historical levels below the normal range but not substantiated in current testing); and 55 individuals that tested within the normal range and did not have historic levels that were low. Full-length VWF sequencing was performed on all index cases and sequence variations (SVs) were identified, and where possible, compared to clinical phenotypes reported in the Sheffield Database. SVs were identified in 53% of the Type 1 VWD cohort (excluding those that were normal on testing without abnormal historic results). Of the 232 Type 1 VWD with VWF levels <40, 74% had SVs. In further analysis of this group, 100% of severe type 1, and 85% of type 1C had SVs. Looking at this in the type 1 with VWF:Ag <40 IU/dL by level of VWF:Ag, 87% of those 1 with VWF:Ag of 2-10, 93% with VWF:Ag 11-20, 71% with VWF:Ag of 21-30, 67% with VWF:Ag of 31-40, and 52% with VWF:Ag of >40 had SVs. The milder phenotypes demonstrate SVs in 39% of the LVWF subjects and 30% of the Type 1H subjects. In the individuals entered into the study as VWD subjects that on central testing had normal levels of VWF on entry into the study and did not have an abnormal historic VWF determination, only 22% had SVs.

A similar approach was undertaken to compare the quantitative BS using the ISTH Bleeding Assessment Tool. For this analysis, a score of 5 or greater was considered abnormal without regard to age or sex. Abnormal BS were identified in 63% of those with VWF:Ag of 2-10, 66% of those 11-20, 64% of those 21-30, 48% of those 31-40, and 58% of those >40. In LVWF subjects, 52% had abnormal BS and in Type 1H subjects 57% were abnormal.

Based upon a similar laboratory analysis, all 2017 family members underwent phenotypic assignment and were found to either be affected or unaffected family members. 713 were phenotyped as being an affected family member and 1296 as unaffected family members. Abnormal bleeding scores were identified in 38% of affected family members and 19% of unaffected family members.

This report summarizes some of the results on the completed Type 1 VWD cohort that was part of the ZPMCBVWD. All 2915 subjects (including index cases, family members, affected family members and normal controls) underwent extensive laboratory testing and phenotype assignment. Of the 499 subjects in the Type 1 VWD cohort, 325 had low VWF and 232 had VWF levels <40 IU/dL. Based on the drop of SV frequency, the level of VWF <40 may provide a possible demarcation to define VWD. In this study the frequency of abnormal bleeding score was not helpful in further defining this critical level for diagnosis - possibly because the bleeding score was ascertained at the entry into the study rather than at the time of diagnosis as done in earlier studies. Further analysis of this cohort can be expected to shed further understanding of the molecular and clinical biology of VWD.