Vonicog Alfa versus Plasma-Derived von Willebrand Factor During Hospitalization: Results of an Observational Retrospective Multicenter Study

Abstract

Background

Available products for the treatment of von Willebrand disease (VWD) now include a new recombinant von Willebrand factor (rVWF) in addition to plasma-derived concentrates (pdVWF). However, these two therapies have never been compared directly in either preclinical studies or real-world inpatient settings.

Objective

The Hopscotch Will II study examined the treatment of VWD in hospitals and compared the use of pdVWF and rVWF.

Methods

Five Rare Bleeding Diseases Centers in Western France retrospectively included patients with VWD over a 48-month period. The data was collected from the BERHLINGO Research Database as well as the French Hospital database, which contains medical information from patient records.

Results

Of the 866 patients evaluated in the study, 285 underwent 648 hospitalizations; 126 adult patients (VWD type 3 excluded) were given VWF concentrates during 249 of those hospital stays. rVWF was used in 61% of the cases. The majority of the hospitalizations were motivated by cutaneous-mucosal symptoms in gastroenterology, stomatology, gynecology, and obstetrics. Consumption of rVWF was lower, though the difference in total VWF consumption per stay or per patient per year was not significant: 51 (57)/34 (30) IU/kg/patient/year for pdVWF versus 40 (47)/27 (26) for rVWF (mean [SD]/median [IQR], Wilcoxon rank sum test, p = 0.2025).

Conclusion

rVWF was used in similar patient profiles and for identical procedures, but the cost of treatment with rVWF was significantly lower, regardless of whether or not FVIII was added.

Keywords

bleeding - hospitalization - inpatients - von Willebrand disease - von Willebrand factor

Von Willebrand disease (VWD) is the most common constitutional hemorrhagic disease in the world, with an estimated prevalence of 1% in the general population and 0.01% for the most severe and symptomatic forms requiring specific treatments.[1] [2] It is characterized by a partial or total quantitative deficiency (VWD type 1 or 3) or a qualitative deficiency (VWD type 2) of von Willebrand factor (VWF), and can affect both women and men.[3] The clinical manifestations of the disease are heterogeneous, mainly consisting of cutaneous-mucosal bleeding[4]: epistaxis, gingivorrhagia, hematoma, menometrorrhagia, etc. Severe bleeds, particularly when related to the gastrointestinal or cerebral system, can lead to direct life-threatening outcomes. Depending on the severity of the VWD, different treatments are available[5]: tranexamic acid and desmopressin (DDAVP) for minor procedures, VWF concentrates with or without factor VIII (FVIII) for invasive procedures or major bleeding. Due to the risk of bleeding, even minor procedures must be supervised through individualized therapeutic protocols, with appropriate clinical and biological monitoring, and most frequently in hospital settings at Rare Bleeding Diseases Centers (RBDCs) in France. In more severe forms, outpatient prophylaxis may also be offered to patients.[6]

A new recombinant VWF (rVWF, vonicog alfa, VEYVONDI [Europe]/VONVENDI [United States], Takeda Pharmaceuticals) has been available in France since December 2018[7] under marketing authorization granted for the prevention and treatment of bleeding in surgical settings in adults only (≥18 years of age). Vonicog alfa complements the therapeutic arsenal of previously available plasma-derived VWF concentrates (pdVWF): highly purified VWF (human VWF, WILFACTIN/WILLFACT/WILSTART (with human FVIII), LFB) or FVIII/VWF combinations at a ratio of 1.0:2.4 (human FVIII and human VWF, VONCENTO, CSL Behring) or a ratio of 1.0:1.0 (human FVIII and human VWF, EQWILATE, OCTAPHARMA). Vonicog alfa is a highly purified rVWF with trace amounts of FVIII; it contains a significant proportion of high molecular weight multimers, which are thought to improve platelet aggregation and collagen binding.[8]

The Hopscotch Will I study (HWI) conducted between 2015 and 2018 described the use of pdVWF concentrates in five French hospitals.[9] The main factors identified as influencing greater consumption of VWF concentrate were the invasive/surgical nature of the procedures performed and the young age of the patients. To evaluate the real-world impact of vonicog alfa in practice, the Hopscotch Will II study proposed a new analysis comparing the use of pdVWF and rVWF concentrates in the same setting.

Methods

Study Design

Hopscotch Will II was an epidemiological, noninterventional, multicenter, open-label, and retrospective study. Five French RBDCs belonging to the BERHLINGO research network participated: Angers, Brest, Le Mans, Nantes, and Rennes. Data from the study were collected over a 48-month period from January 2019 to December 2022. The study is registered at www.clinicaltrials.gov under the number NCT04887324 (May 14, 2021).

Study Objectives

The primary objective of the study was to describe the overall VWF concentrate consumption of treated inpatients with VWD (per patient per year).

The secondary objectives were:

• – To describe the characteristics of VWD patient populations during the study period, and especially the population hospitalized and treated with VWF (type/subtypes of VWD, basal biomarkers).

• – To describe and compare treatment regimens during hospitalization in different subgroups (VWD type and subtypes, type of VWF treatment, reasons for hospitalization).

• – To evaluate the costs of hospital treatment with pdVWF and rVWF per patient per year from the perspective of the French health insurance system (at the national responsibility price, as of December 31, 2022). These tariffs remained unchanged over the 4-year study (2019–2022), for both VWF and FVIII specialties.

Data Sources and Data Collection

Data were collected retrospectively from two different sources:

• – The BERHLINGO hemostasis research database of Western France, which contains de-anonymized information such as clinical-biological characteristics and patient-specific treatment modalities.

• – The French Medical Information System (Programme de Médicalisation des Systèmes d'Information = PMSI), which contains data about hospital stays, such as the major diagnostic categories (MDCs) with their diagnostic codes and the medical procedures performed (surgical, imaging, surveillance, etc.), and expensive drugs provided.

Population

Characteristics. The eligible patients had a constitutional VWD and met the same inclusion criteria as for the HWI study[9]: VWD type 3 (severe form with VWF antigen (VWF:Ag) and VWF activity (VWF:Act) < 5 IU/dL or %) and VWD type 2 (2A, 2B, 2M, or 2N). In the absence of the above-mentioned criteria, we chose to include patients with VWF:Act < 30 IU/dL (especially for VWD type 1), given that molecular biological analyses are not provided to all patients.

Patients under 18 years of age were considered pediatric patients. Sex was the sex assigned at birth. Patients under guardianship were excluded.

Patients with VWF treatment. Since the PMSI can only identify expensive drugs (tarification à l'activité T2A, in French), only patients who received clotting factors (CF) during hospitalization were considered “treated.” Patients receiving DDAVP and/or tranexamic acid or not receiving hemostatic therapy were classified as “untreated.” For treated patients, the data of pdVWFs were grouped, regardless of their FVIII/VWF ratio or their multimeric profile. Indeed, the French recommendations (Protocole National de Diagnostic et de Soins PNDS) indicate that the efficacy data of the different VWF concentrates are equivalent and do not allow prioritizing the use of one type of concentrate over another[10]; in addition, due to the obligations of compliance with the public procurement code and prolonged interruptions in the availability of certain specialties, physicians are used to modifying their patient's treatment with pdVWF and moving from one specialty to another. Because rVWF is only approved in France for use in patients aged 18 years and older, we also focused on this specific population. Furthermore, rVWF does not have authorization for prophylaxis in France; thus, we chose to present the outcomes of patients with VWD type 3 in a distinct manner, since this could have resulted in a smaller switch toward rVWF in this population.

Bleeding. The term “bleeding” was retained when a diagnostic code or a CCAM (Common Classification of Medical Procedures) act code referred to bleeding (hematoma, hemorrhage, melena, etc.). In all cases, a hemorrhagic episode is present at the end of the stay (upon admission or during the stay), but in view of the codes, it is not always possible to distinguish the bleeding that is the cause of hospitalization from the bleeding resulting from a surgical procedure, for example. Nevertheless, considering the other diagnostic codes and other CCAM act codes present, massive bleeding refers to hemorrhagic episodes present at admission (>95%).

Statistical Analysis Plan

No inferential analysis involving formal testing was planned in this noncontrolled study. Every patient who agreed to participate in the study was included. Therefore, the recruitment was as exhaustive as possible. Missing data were not imputed: calculations were based on observed values only.

Patient characteristics and/or administrations of VWF and/or FVIII concentrates were described comprehensively and by subgroup. Quantitative data were expressed as a median (interquartile range, IQR) and/or mean (standard deviation, SD) and/or minimum/maximum (min/max). Categorical data were presented as numbers and/or percentages. Statistical analyses were performed with the Pearson's chi-square test, the Fisher's exact test, the Kruskal–Wallis rank sum test, and the Wilcoxon rank sum test. The significance level was set at 5%. The effect of dependent variables on VWF consumption was determined through analysis of variance.

All analyses were performed with the statistical software Jamovi 2.3.28 (using R version 4.4.1).

Ethical Considerations

The BERHLINGO research database meets the requirements of the French Data Protection Committee (CNIL, “reference methodology”—Méthodologie de référence MR004). The Institutional Ethics Committee approved the protocol on July 7, 2021 (Groupe Nantais d'Ethique dans le Domaine de la Santé, GNEDS).

All patients eligible for inclusion in the study were sent a written information sheet (and were allowed a right of refusal). Patients who objected to participating in the study were excluded; patients who did not object were considered included in the study. Written consent was not required, as per the reference methodology MR004.

The study report is in line with the general guidelines of the EQUATOR network, and in particular the STROBE methodology.[11]

Results

A total of 1,214 patients were screened. After deleting the data of patients who met one of the exclusion criteria, eliminating duplicates or patients lost to follow-up (n = 324 patients) and collecting patient objections to participate (n = 24 patients), the data of 866 patients were analyzed in the Hopscotch Will II (HWII) study over the 2019 to 2022 period (see Supplementary Material S1, available in the online version only).

Characteristics of Patients in the Cohort

Whole cohort. The characteristics of the population are presented in Supplementary Material S2 (available in the online version only; n = 866 patients). The population area covered by the five RBDCs on January 1, 2021, included about 7,248,566 inhabitants.[12] [13] The patients with VWD living in this area represented 0.012% of the population, that is, 12.0 people with VWD/100,000 inhabitants.

For information, one patient had allo-immunised type 3 VWD. In 2019, she began prophylactic treatment with emicizumab and was treated solely with high doses of FVIII during her hospitalization. VWF was never given to her. Despite being included in the study, this patient was not included in the VWF consumption analysis for this reason.

VWF Treatments in Adult Inpatients without Inhibitors (VWD Type 3 Excluded)

The use of rVWF in hospitalization gradually emerged in Western France to reach 73% of consumption in 2022, mainly at the expense of VWF/FVIII mixes (see [Table 1]). The focus was on hospital stays where adult patients were exclusively treated with pdVWF or rVWF (with the exception of three short-term stays where patients were given a single or two injections of each type of VWF; see [Fig. 1]).

Fig. 1 Flowchart of hospital stays by treatment for adult inpatients treated with VWF and/or FVIII. Depicts the distribution of hospital stays for adult patients by treatment administered for adult inpatients. ^*Nine patients (who went from minor to adult during the study period) were hospitalized as patients < 18 years over the 4-year study. CF, clotting factors; FVIII, factor VIII; pdVWF, plasma-derived von Willebrand factor; RBDC, rare bleeding diseases center; rVWF, recombinant von Willebrand factor.

Table 1 Part of pdVWFs versus rVWF in hospitalization per year (% of total IU): HWII study versus France

		2019	2020	2021	2022
pdVWF/FVIII	HWII	0%	12%	8%	5%
(Ratio 1:1)	France	1%	2%	4%	5%
pdVWF/FVIII	HWII	45%	47%	16%	20%
(Ratio 1:2.4)	France	14%	23%	8%	10%
pdVWF (highly purified)	HWII	31%	5%	3%	2%
	France	81%	66%	70%	68%
rVWF	HWII	24%	36%	73%	73%
	France	4%	8%	18%	17%

Abbreviations: ATIH, Agence technique de l'information sur l'hospitalisation (Technical Agency for Information on Hospital Care); FVIII, factor VIII; HWII, Hopscotch Will II study; IU, international unit; MSO, Medical surgical and obstetrics; pdVWF, plasma-derived von Willebrand Factor; rVWF, recombinant von Willebrand Factor; UCD, unité de dispensation commune (code of dispensing unit).

Note: Variables are described as percentages. National consumptions were extracted from the ATIH database: expensive drugs consumptions by UCD code, public sector. Field: MSO. Accessed June 12, 2003 at: https://www.scansante.fr/applications/synthese-dmi-mo-sus.

Characteristics of adult patients treated with VWF (VWD type 3 excluded). In total, 126 adult patients were treated with VWF over 249 stays. The mean baseline factor levels (SD) excluding VWD type 2N were 31 (21) IU/dL for VWF:Ag, 15 (11) IU/dL for VWF:Act, and 45 (21) IU/dL for basal FVIII (FVIII:C). Thrombocytopenia was present in 4.0% of the patients, 60% of whom had VWD type 2B. There was a slight overrepresentation of patients with type O blood type (49% vs. 42% in the Caucasian population[14]); this proportion rose to 59% in patients with VWD type 1. Regarding the VWF genetic defects represented, 64.7% of the patients were genotyped, 11.8% were not genotyped but had a family history, and 23.5% were not genotyped and had no known family history.

[Table 2] describes the characteristics of adult patients treated at least once with pdVWF versus patients treated at least once with rVWF.

Table 2 Characteristics of adult inpatients treated with rVWF versus pdVWF (VWD type 3 excluded)

	VWF (126 inpatients)	pdVWF at least once (60 inpatients[a])	rVWF at least once (85 inpatients[a])	p-Value (pdVWF vs. rVWF)
Sex (male/female)	46/80 (37%/63%)	25/35 (42%/58%)	26/59 (31%/69%)	0.1688, ns
Weight (kg)	71.5 (16.8)/67.0 (23.0)	73.2 (18.8)/68.8 (20.8)	67.5 (15.8)/65.0 (19.1)	0.5244, ns
Age at admission (y)	49.3 (18.6)/49.4 (30.1)	49.9 (19.4)/48.3 (29.4)	53.9 (17.5)/52.0 (30.0)	0.3404, ns
Blood group	116 (92%)	57 (95%)	78 (92%)	0.4527, ns
O	57 (49%)	26 (46%)	38 (48%)
A	51 (44%)	28 (49%)	34 (44%)
AB	1 (1%)	1 (2%)	0 (0%)
B	7 (6%)	2 (3%)	6 (8%)
Basal plasmatic levels
VWF:Ag (IU/dL)	32.0 (22.4)/25.0 (21.5)	25.8 (14.9)/24.0 (18.9)	30.6 (24.4)/29.0 (23.0)	0.0122, significant
VWF:Act (IU/dL)	16.9 (16.4)/14.0 (11.0)	13.9 (8.0)/12.0 (7.3)	14.8 (15.6)/14.0 (10.1)	0.1737, ns
FVIII:C < 40 IU/dL	57 (46%)	31 (52%)	34 (40%)	0.1641, ns
VWD type
NC	3 (2.4%)	2 (3.3%)	2 (2.3%)	0.3191 (types of VWD), ns 0.2755 (subtypes of VWD type 2), ns
Type 1	34 (27.0%)	19 (31.7%)	18 (21.3%)
Type 2	89 (70.6%)	39 (65.0%)	65 (76.4%)
VWD2A	26 (20.6%)	10 (16.7%)	21 (24.7%)
VWD2B	16 (12.7%)	4 (6.7%)	15 (17.6%)
VWD2M	32 (25.4%)	19 (31.6%)	19 (22.4%)
VWD2N	3 (2.4%)	1 (1.7%)	2 (2.3%)
VWD2NC	12 (9.5%)	5 (8.3%)	8 (9.4%)
Type of outpatient treatment
On-demand treatment	126 (100%)	60 (100%)	85 (100%)	NA
Characteristics of stays
Number of stays/patient/y	0.5 (0.6)/0.2 (0.2)	0.4 (0.2)/0.2 (0.2)	0.5 (0.6)/0.2 (0.2)	0.2591, ns

Abbreviations: FVIII:C, plasma factor VIII assay; kg, kilogram; n, number; NA, not applicable; NC, not classified; ns, not significant; pdVWF, plasma-derived von Willebrand Factor; rVWF, recombinant von Willebrand Factor; VWD, von Willebrand disease; VWF:Act, plasma von Willebrand Factor assay (Activity); VWF:Ag, plasma von Willebrand Factor assay (Antigen).

Note: Qualitative variables are described as numbers and/or percentages; they were compared using the Pearson's chi-square test or the Fisher's exact test (if effective < 5), α = 0.05. Quantitative variables are described as a mean (standard deviation, SD) and median (interquartile range, IQR), and compared using the Wilcoxon rank sum test, α = 0.05.

^a A same patient may have been treated with pdVWF and rVWF during the 4-year study (n = 18/126 patients = 14.3%).

Primary causes of hospitalization with VWF treatment (VWD type 3 excluded). As seen in [Fig. 2], the causes mainly concerned the following clinical areas:

Fig. 2 Characteristics of stays according to age range for adult inpatients treated with pdVWF or rVWF. Depicts the evolution of the characteristics of hospital stays over time, according to age group for adult inpatients treated with pdVWF or rVWF. Sessions: the term hospital “sessions” refers to a specific major diagnostic category, called “CMD-28 Séances,” referring to very short-term stays that could be similar to dialysis sessions, for example. In these types of stays, it is particularly possible to administer very specific therapeutics (such as coagulation factors, blood transfusions, IV iron, DDAVP, etc.). Miscellaneous: diseases of the respiratory system, hematological malignancies, etc. Circ., circulatory system; gyn.-obst., gynecology-obstetrics; HGE, hepato-gastroenterology; H&N, head and neck; misc, miscellaneous; nb, number; nerv., nervous system; trauma/ortho, traumatology/orthopedics; VWF, von Willebrand factor.

• – Hepato-Gastroenterology for 26% of the stays, essentially for digestive explorations.

• – Gynecology and obstetrics for 22% of the stays, mainly for deliveries and in vitro fertilization procedures.

• – Head and Neck for 17% of the stays, mainly for dental avulsions.

• – Traumatology for 11% of the stays (57% of which are for major surgeries such as prostheses/arthrodesis, etc.).

Active bleeding was involved in 26% of the hospital stays, with melena playing a major role in 48% of cases.

VWF concentrates use (VWD type 3 excluded). Over the 2019 to 2022 period, the 249 stays with the administration of VWF can be divided into two main categories depending on the type of concentrate used:

• – Thirty-nine percent of stays with pdVWF (97 stays/60 patients): pdVWF alone (34 stays/21 patients), pdVWF and FVIII (2 stays/2 patients), and combined pdVWF/FVIII (61 stays/39 patients).

• – Sixty-one percent of stays with rVWF (152 stays/85 patients): rVWF alone (123 stays/73 patients), rVWF and FVIII (29 stays/16 patients).

Consumption of VWF and FVIII concentrates during hospital stays (adult patients without inhibitors, VWD type 3 excluded). To circumvent the notion of weight, the results were rendered in VWF IU/kg/patient/year (and in VWF IU/kg/stay in Supplementary Material S3, available in the online version only).

A reduction of consumption was observed with rVWF, and was significant for VWD type 1 (pdVWF vs. rVWF):

^*VWD type 1: p = 0.0103:

• – Mean (SD) consumption of 42 (36) versus 19 (11) IU/kg/patient/year.

• – Median (IQR) consumption of 36 (27) versus 18 (11) IU/kg/patient/year.

• ^*VWD type 2: p = 0.3122:

• – Mean (SD) consumption of 60 (67) versus 41 (39) IU/kg/patient/year.

• – Median (IQR) consumption of 31 (41) versus 30 (30) IU/kg/patient/year.

Invasive and noninvasive procedures have a different distribution, with an over-representation of noninvasive procedures treated by rVWF (chi-square p = 0.018, Cramér's V = 0.163). The frequency of minor versus major procedures was comparable for the two types of VWF (p = 0.1009). The proportion of bleeding rates was also similar between noninvasive and invasive procedures (p = 0.0980); the proportion of bleeding rates was higher in major invasive procedures than in minor ones (p = 0.018). There were no significant differences in consumption for any type of procedures, with or without bleeding (see [Table 3]). An analysis of variance did not show a significant difference in VWF consumption, depending on the type of procedure and the VWF type, substantially (invasive vs. noninvasive procedures: p = 0.1014).

Table 3 Consumption of pdVWF versus rVWF in adult inpatients (IU/kg/patient/year) according to the type of procedure, the presence or absence of bleeding (n = 126 patients)

	pdVWF (n = 60 patients[a])	rVWF (n = 85 patients[a])	p-Value
Noninvasive procedures	22 (14)/18 (13)	36 (61)/19 (27)	0.7493, ns
No bleed	18 (12)/14 (14)	34 (54)/14 (26)	0.6470, ns
Bleeds	21 (13)/17 (15)	29 (26)/17 (29)	0.7616, ns
Invasive procedures	54 (60)/36 (43)	36 (32)/26 (23)	0.2453, ns
Major	45 (50)/30 (26)	33 (23)/28 (21)	0.6982, ns
No bleed	37 (43)/27 (20)	31 (15)/29 (17)	0.5849, ns
Bleeds	81 (70)/62 (84)	31 (17)/28 (15)	0.2065, ns
Minor	47 (57)/25 (40)	31 (26)/21 (25)	0.7028, ns
No bleed	21 (12)/18 (19)	23 (15)/19 (18)	0.7919, ns
Bleeds	94 (71)/83 (96)	41 (33)/37 (57)	0.0903, ns
Total	51 (57)/34 (30)	40 (47)/27 (26)	0.2025, ns

Abbreviations: FVIII:C, plasma factor VIII assay; kg, kilogram; ns, not significant; pdVWF, plasma-derived von Willebrand Factor; rVWF, recombinant von Willebrand Factor.

Note: Quantitative variables are described as a mean (standard deviation, SD) and median (interquartile range, IQR). They were compared using the Wilcoxon rank sum test, α = 0.05.

^a Sixteen percent of patients have been treated with either pdVWF or rVWF during their various stays.

No significant difference between pdVWF and rVWF was noted for VWF consumption in IU/kg/patient/year in the main clinical areas (hepato-gastroenterology p = 0.2337/head and neck p = 0.9424/gynecology-obstetrics p = 0.1492/traumatology p = 0.5002).

For example, for the subsequent types of stays (mean [SD] pdVWF vs. rVWF consumption):

• – Vaginal deliveries with suture (major procedures): 141 (118) versus 155 (71) IU/kg/stay, for an average length of stay of 6.0 (0.0) versus 4.8 (1.5) days of hospitalization.

• – Dental avulsions (minor procedures): 69 (46) versus 64 (26) IU/kg/stay, for an average length of stay of 1.3 (0.5) versus 1.5 (0.5) days of hospitalization.

–Melena with simple digestive explorations such as colonoscopies/videocapsules (minor procedures): 223 (145) versus 151 (87) IU/kg/stay, for an average length of stay of 7.0 (5.7) versus 7.3 (2.7) days of hospitalization.

In the end, no significant difference was found between the use of pdVWF and rVWF treatments, despite a downward trend in favor of rVWF (pdVWF vs. rVWF, p = 0.2025):

• – Mean consumption (SD) of 51 (57) versus 40 (47) IU/kg/patient/year.

• – Median consumption (IQR) of 34 (30) versus 27 (26) IU/kg/patient/year.

The median (IQR) length of stay was similar between pdVWF and rVWF: 2.0 (4.0) days (p = 0.7051). Moreover, the median dose of VWF per day of hospitalization was similar between pdVWF and rVWF: 38 (22) versus 32 (26) IU VWF/kg/day of hospitalization (p = 0.0750).

The four specialties of VWF have no significant differences in annual VWF consumption per patient per year when compared (p = 0.0825; see Supplementary Material S4, available in the online version only).

Moreover, the annualized pdVWF consumption for adult treated patients, both included in HWI and HWII studies, is consistent across both studies (p = 0.7922):

• – Exclusion of patients with VWD type 3, n = 106 patients (80 patients treated with pdVWF during HWI 2015–2018 and 46 patients during HWII 2019–2022).

• – Mean (SD) consumption HWI versus HWII = 56 (105) versus 52 (60) IU/kg/patient/year.

• – Median (IQR) consumption HWI versus HWII = 32 (35) versus 34 (33) IU/kg/patient/year.

Cost of VWF Treatment during Hospitalization (Adult Patients without Inhibitors, VWD Type 3 Excluded)

From the perspective of the French health insurance system, rVWF has a significantly lower annual cost per treated inpatient—including the additional cost of FVIII—(p = 0.0421):

• – Mean (SD) cost of €3,595 (4,195) versus €2,620 (2,915).

• – Median (IQR) cost of €2,414 (2,877) versus €1,834 (1,977).

The cost per patient per stay was also significantly lower with rVWF, regardless of the addition of FVIII (p = 0.0056).

VWF Consumption in VWD Type 3 (Adult Patients without Inhibitors)

Patients with VWD type 3 without inhibitors (n = 7) were treated with VWF during 14 stays:

• – Six patients received pdVWF during 10 stays.

• – Three patients received rVWF during four stays.

Half of the stays were noninvasive procedures, 29% were minor invasive procedures, and 21% were major invasive procedures. Stays in HGE and in traumatology/orthopedics each accounted for 36% of hospitalizations, including 60% of melena for HGE and 40% of hemarthrosis for traumatology.

The mean consumption (SD) was 60 (42) versus 25 (26) IU/kg/patient/year, and the median consumption (IQR) was 48 (26) versus 11 (23) IU/kg/patient/year (pdVWF vs. rVWF).

Discussion

The rationale for the development of rVWF was to reshape the treatment landscape for VWD globally, by producing a highly purified VWF for use in combination with FVIII if necessary—especially in countries where only VWF/FVIII combinations had been available previously. We conducted the first French study matching validated clinical-biological data to real-life use of rVWF in hospitals compared to pdVWF.

At the epidemiological level, we noted a higher than expected prevalence of VWD in our whole cohort: 12.0 patients/100,000 inhabitants compared to only 5.3 patients/100,000 inhabitants in the French national registry or 5.4 patients/100,000 inhabitants in the Italian registry.[15] [16] Nevertheless, this result was very close to the prevalence reported in the United Kingdom in 2020, which was 16.5 patients/100,000 inhabitants.[17] Furthermore, the clinical and biological data (VWD type, basal levels) were consistent with those reported in 2016 by Veyradier et al,[18] but with a slightly lower proportion of VWD type 2B (11% vs. 17%) and of VWD type 3 (3% vs. 8%) in our cohort.

The most prevalent clinical areas during hospital stays were those expected based on the pathophysiology of the disease: the digestive, gynecological-obstetric, and head and neck domains in particular.[19] [20] We confirmed a trend toward lower consumption of rVWF (overall, for most VWD types/subtypes, by type of procedure, by clinical area) and coupled with significantly lower overall costs. Due to the heterogeneity of the populations treated and the variety of ways in which VWF consumption is expressed in the literature (per injection, per stage of the surgical procedures, per year, etc.), it is difficult to compare our results with those of other teams. Furthermore, no publication to date has compared rVWF consumption with pdVWF consumption. We found a median consumption of 64 IU/kg/stay for rVWF that rose to 89 IU/kg/stay for invasive surgeries/procedures (114 for major surgeries and 58 for minor surgeries). Our doses of rVWF were much lower than those found in the clinical study published in 2019 by Peyvandi[21]: 220 IU/kg per surgical stay on average, ranging from 120 to 308 IU/kg/stay between minor and major surgeries. This could be explained by the large proportion of patients with VWD type 3 included in research protocols (in the 2019 Peyvandi study: 53% of patients had VWD type 3), whereas the patients with VWD type 3 were excluded from our analysis. In France, prophylaxis with rVWF is not allowed, and the patients concerned, especially those with VWD type 3, are usually left on pdVWF during their hospital stay. As for populations recruited in studies throughout France or nearly throughout France in 2021, Desprez et al described a set of 63 stays under treatment with rVWF[22]: the median doses were lower than those in our study for major surgeries (63 vs. 114), minor surgeries (37 vs. 58) and overall for invasive procedures (63 vs. 89). This may be explained by the high proportion of patients with VWD type 1 (40%), who generally require lower treatment doses, compared to our study (only 11% VWD type 1 in surgery under rVWF). On the other hand, Sun et al described a cohort in 2023 with 79.1% of patients recruited in France[23]: the mean reported dose of rVWF treatment for surgery of 105 IU/kg/stay was close to the mean dose we report (89 IU/kg/stay for invasive procedures). Nonetheless, we clearly observed a reduction in costs with rVWF compared with pdVWF, for similar reasons for hospitalization and similar consumption levels. FVIII/VWF combinations are intrinsically more expensive than highly purified VWF; the cost reduction was seen despite the possible additional cost of FVIII at the initiation of treatment.

This study should nevertheless be interpreted with caution due to a few limitations. The difficulty in categorizing invasive procedures as involving a minor or major risk of bleeding was a challenge from a methodological perspective, as there is no national or international consensus on this issue in VWD. A few classifications have been proposed, but they are partial or nonspecific to VWD.[24] [25] Although a difference in the distribution of procedures was shown between pdVWF and rVWF (for noninvasive procedures/not for other types of procedures nor bleeding), the effect is small (Cramér's V ≤ 0.200), and an ANOVA did not show a significant difference between the consumptions. In total, this difference is minor and does not alter the main result: the overall consumption of rVWF remains lower for all hospital stays combined, and the observed effect of the types of procedures has a negligible impact on this conclusion. Furthermore, this work only addresses the evaluation of hospital therapeutic care for a limited period, and the figures obtained cannot be compared to modeling data that considers the entire lifespan of a patient, for instance. The consistency of product use in IU/kg/yr across HWI and HWII is reassuring, though.

As expected, French practitioners have a good grasp of the use of a highly purified VWF in real life and, in particular, of whether or not to add FVIII at the beginning of treatment. Worldwide, this necessity could be perceived as a drawback by physicians unfamiliar with this use. While Peyvandi et al reported supplementation with FVIII in 30% of major invasive surgeries, we only found the addition of FVIII in 18.0% of stays for major surgery under rVWF. This difference was certainly explained by the absence of VWF injections 12 to 24 hours before surgery as preoperative prophylaxis at our center: injections did not begin until the day of the procedure.

In the end, there was no significant difference in VWF consumption, though we did observe a general downward trend in favor of rVWF. However, we were not able to identify any specific population as more likely to benefit from the administration of rVWF instead of plasma derivatives. At the time of the study, our database was limited to the total amount of VWF administered during hospitalization; thus, we were not able to determine whether surgical patients received equivalent doses of pdVWF or rVWF for a set period of time after discharge. The impact on the total consumption of packages and, in particular, the number of IU per vial, which differ between different products, was not evaluated either. Our study did not show a difference between the two types of VWF, supporting the idea that use is similar for the same types of patients, procedures and clinical areas, but our analysis may not have revealed a factual difference due to the limited number of study participants, the difficulties of conducting subgroup analyses in this small population, and the outsized impact of potential outliers.

Conclusion

The HWII study provides an initial glimpse at the actual usage of rVWF in hospitals and compares it to plasma derivatives. The data collected thus far indicate that the different products are equivalent, with a trend toward lower consumption with rVWF at a lower price. rVWF constitutes a new alternative to existing therapies, without the superfluous addition of FVIII. Nevertheless, further in-depth analyses at a larger scale will be required to confirm and refine these results.

Conflict of Interest

A.B. received funding from Takeda. M.F. received funding from Takeda and consulting fees from CSL Behring. C.T. received consulting fees from LFB. B.G. received research grants and consulting fees from CSL Behring. M.T. received funding from Baxalta GmbH for the Hopscotch Will II study (via a research grant awarded to Nantes University Hospital).

Acknowledgment

The authors thank the patients and their families for participating in this study. The authors would like to thank the clinical trial staff at each of the centers for their meticulous work and the medical information departments for their precious contribution to the results of this study. This study was conducted within the French HUGO network of University Hospitals (Hôpitaux Universitaires du Grand Ouest). The authors would also like to acknowledge Ms. Jill Rupnow (Kokopelli SARL, Paris, France) for providing medical writing services; these services were funded through the study budget.

Data Availability Statement

The Nantes University Hospital does not plan to share data supporting the results reported in this article.

Contributors' Statement

All authors contributed to the design of the study, the acquisition of data, and the interpretation of data. V.H. and M.T. contributed to the conception of the study and the analysis of data. All authors commented on all intermediary versions of the manuscript (drafting the article or revising it critically for important intellectual content). All authors read and approved the final version of the manuscript to be submitted.

Clinical Trial Registration

The study is registered at www.clinicaltrials.gov under the number NCT04887324 (May 14, 2021).

• References

• 1 Sharma R, Flood VH. Advances in the diagnosis and treatment of von Willebrand disease. Blood 2017; 130 (22) 2386-2391
• 2 Bowman M, Hopman WM, Rapson D, Lillicrap D, James P. The prevalence of symptomatic von Willebrand disease in primary care practice. J Thromb Haemost 2010; 8 (01) 213-216
• 3 Sadler JE, Mannucci PM, Berntorp E. et al. Impact, diagnosis and treatment of von Willebrand disease. Thromb Haemost 2000; 84 (02) 160-174
• 4 Castaman G, Linari S. Diagnosis and treatment of von Willebrand disease and rare bleeding disorders. J Clin Med 2017; 6 (04) 45
• 5 Peyvandi F, Kouides P, Turecek PL, Dow E, Berntorp E. Evolution of replacement therapy for von Willebrand disease: from plasma fraction to recombinant von Willebrand factor. Blood Rev 2019; 38: 100572
• 6 El Alayli A, Brignardello Petersen R, Husainat NM. et al. Outcomes of long-term von Willebrand factor prophylaxis use in von Willebrand disease: a systematic literature review. Haemophilia 2022; 28 (03) 373-387
• 7 Ministère des Solidarités et de la Santé. Arrêté du 14 décembre 2018 modifiant la liste des spécialités pharmaceutiques agréées à l'usage des collectivités et divers services publics. 2018 . Accessed September 15, 2025 at: https://www.legifrance.gouv.fr/download/pdf?id=o4_gmvmogk5ztbgjxvtza9fqnu3rol51x4_ogbphbgy=
• 8 Franchini M, Mannucci PM. Von Willebrand factor (Vonvendi): the first recombinant product licensed for the treatment of von Willebrand disease. Expert Rev Hematol 2016; 9 (09) 825-830
• 9 Horvais V, Beurrier P, Cussac V. et al; BERHLINGO Consortium. Key drivers of coagulation factor use in von Willebrand disease during hospitalization: an overview of the French BERHLINGO cohort. Clin Drug Investig 2024; 44 (01) 35-49
• 10 CRMW. Protocole National de Diagnostic et de Soins (PNDS): Maladie de Willebrand. HAS; 2021 . Accessed January 2, 2026 at: https://www.has-sante.fr/upload/docs/application/pdf/2021-02/maladie_de_willebrand_-_argumentaire.pdf
• 11 von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP. STROBE Initiative. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet 2007; 370 (9596) 1453-1457
• 12 INSEE. TCRED - Estimations de populations - Ensemble - Loire-Atlantique. 2023 . Accessed February 22, 2024 at: https://www.insee.fr/fr/statistiques/serie/001760124
• 13 INSEE. TCRED - Estimations de Population - Ensemble - Vendée. 2023 . Accessed February 22, 2024 at: https://www.insee.fr/fr/statistiques/serie/001760165
• 14 Vallois HV, Marquer P. La répartition en France des groupes sanguins ABO. Bull Mem Soc Anthropol Paris 1964; 6 (01) 1-200
• 15 FranceCoag. Données démographiques: Maladie de Willebrand. Données au 27/11/ 2023 . FranceCoag. November 27, 2023. Accessed February 27, 2024 at: https://www.francecoag.org/sitewebpublic/public/stats/stats_page.jsp?stat4=on
• 16 Abbonizio F, Hassan HJ, Riccioni R, Santagostino E, Arcieri R, Giampaolo A. Italian Association of Haemophilia Centres (see Appendix). New data from the Italian National Register of Congenital Coagulopathies, 2016 Annual Survey. Blood Transfus 2020; 18 (01) 58-66
• 17 Du P, Bergamasco A, Moride Y, Truong Berthoz F, Özen G, Tzivelekis S. Von Willebrand disease epidemiology, burden of illness and management: a systematic review. J Blood Med 2023; 14: 189-208
• 18 Veyradier A, Boisseau P, Fressinaud E. et al; French Reference Center for von Willebrand disease. A laboratory phenotype/genotype correlation of 1167 French patients from 670 families with von Willebrand disease: a new epidemiologic picture. Medicine (Baltimore) 2016; 95 (11) e3038
• 19 Leebeek FWG, Eikenboom JCJ. Von Willebrand's Disease. N Engl J Med 2016; 375 (21) 2067-2080
• 20 Brignardello-Petersen R, El Alayli A, Husainat N. et al. Surgical management of patients with von Willebrand disease: summary of 2 systematic reviews of the literature. Blood Adv 2022; 6 (01) 121-128
• 21 Peyvandi F, Mamaev A, Wang J-D. et al. Phase 3 study of recombinant von Willebrand factor in patients with severe von Willebrand disease who are undergoing elective surgery. J Thromb Haemost 2019; 17 (01) 52-62
• 22 Desprez D, Drillaud N, Flaujac C. et al. Efficacy and safety of a recombinant Von Willebrand Factor treatment in patients with inherited Von Willebrand Disease requiring surgical procedures. Haemophilia 2021; 27 (02) 270-276
• 23 Sun SX, Lowndes S, Willock R, Jones C, Brighton S. Outcomes in patients with von Willebrand disease receiving recombinant von Willebrand factor on demand and in surgical settings: chart review. Clin Appl Thromb Hemost 2023; 29
• 24 Schulman S, Kearon C. ; Subcommittee on Control of Anticoagulation of the Scientific and Standardization Committee of the International Society on Thrombosis and Haemostasis. Definition of major bleeding in clinical investigations of antihemostatic medicinal products in non-surgical patients. J Thromb Haemost 2005; 3 (04) 692-694
• 25 Solimeno LP, Escobar MA, Krassova S, Seremetis S. Major and minor classifications for surgery in people with hemophilia: a literature review. Clin Appl Thromb Hemost 2018; 24 (04) 549-559

Publication History

Received: 15 July 2025

Accepted: 21 October 2025

Article published online:
13 January 2026