Impact of tranexamic acid on postpartum hemorrhage in type 1 von Willebrand disease treated with recombinant VWF Open Access

Von Willebrand disease (VWD) is the most common inherited bleeding disorder, caused by quantitative or qualitative deficiency of von Willebrand factor (VWF) and affecting ∼1% of the general population.¹ Individuals with VWD experience primarily mucocutaneous bleeding including epistaxis and heavy menstrual bleeding and have an increased risk of bleeding during surgery and childbirth.^2,3 Pregnant individuals with VWD are particularly vulnerable to both primary and secondary postpartum hemorrhage (PPH), with rates up to 44%, significantly higher than in the general obstetric population.^4-8 

PPH remains a major cause of maternal morbidity and mortality worldwide, with consequences extending well beyond acute blood loss in the immediate postpartum period.^9,10 Primary PPH is defined as >1000 mL or hemodynamic instability due to bleeding, which occurs within 24 hours of delivery, and secondary PPH is bleeding from 24 hours to 6 weeks postpartum.^11,12 Women with VWD who experience excess bleeding at delivery have a higher risk of prolonged hospitalization, iron deficiency anemia, impaired lactation, and postpartum depression.^13,14 These maternal outcomes underscore the importance of achieving optimal hemostatic control at delivery to prevent immediate and long-term complications.

Although VWF levels physiologically increase during pregnancy, this rise is often inadequate in women with VWD. International guidelines recommend maintaining VWF and factor VIII clotting activity (FVIII:C) levels >50% to permit safe epidural analgesia and minimize the risk of PPH.^15-18 However, this target may not sufficiently mimic the physiological hemostatic changes observed in women without VWD, who experience up to a 300% elevation in VWF levels at term.^19,20 Such levels are seldom achieved in women with VWD, even with VWF replacement therapy.^21,22 Consequently, there is increasing debate surrounding optimal VWF dosing targets at delivery, with some evidence suggesting that higher replacement levels may improve outcomes.²³ A recent single-institution cohort study reported that targeting higher VWF and FVIII:C levels at delivery (≥100 IU/dL) led to improved hemostatic outcomes in women with VWD, including reduced estimated blood loss, fewer transfusions, and a lower incidence of PPH.²⁴ 

Tranexamic acid (TXA), an antifibrinolytic agent that inhibits the breakdown of fibrin clots and thereby stabilizes blood clots and reduces excessive bleeding, has been recognized for its efficacy in managing PPH in the general population.^25-27 The landmark WOMAN trial demonstrated a significant reduction in PPH-associated mortality when TXA was administered within 3 hours of delivery, solidifying its role as a standard adjunct in PPH management.²⁸ TXA has also been shown to effectively manage heavy menstrual bleeding in women, including those with VWD.^29,30 Although often used in clinical practice, there are limited randomized trial data regarding the efficacy and safety of the combination of TXA with VWF concentrate to optimize hemostasis in women with VWD at delivery.

In this pilot study, we aimed to evaluate the safety and feasibility of administering TXA in combination with recombinant VWF (rVWF) compared with rVWF alone for the prevention of PPH in women with VWD. We aimed to inform future strategies for optimizing hemostatic management in this high-risk population by addressing this critical gap in the evidence base.

VWD-Woman was an investigator-initiated, industry-sponsored, phase 3, open-label, randomized, single-center pilot trial. Participants were screened and enrolled during routine clinic visits at the Hemophilia Center of Western Pennsylvania, with delivery planned at the University of Pittsburgh Medical Center, Magee-Womens Hospital in Pittsburgh, PA. The study was approved by the University of Pittsburgh Biomedical institutional review board, 28 April 2021, and 20 participants were enrolled between 4 June 2021 and 17 May 2024.

The study enrolled pregnant females aged ≥18 years with confirmed VWD, defined by a historical von Willebrand Factor Ristocetin Cofactor activity (VWF:RCo) level of <0.50 IU/mL and a documented bleeding history, who planned delivery at University of Pittsburgh Medical Center, Magee-Womens Hospital. Bleeding history was assessed using International Society on Thrombosis and Haemostasis bleeding assessment tool scores previously completed as part of standard clinical care and documented in the medical record, along with records of previous treatment for mucocutaneous or procedural bleeding episodes. Although individuals with all VWD subtypes were eligible, all enrolled participants had type 1 VWD. Participants were required to consent to blood draws, comply with random assignment, maintain a 21-day postpartum pictorial blood assessment chart (PBAC) diary, document hemostatic agent use, and attend study visits. The 21-day postpartum period was selected to capture secondary outcomes, including delayed or prolonged bleeding, hemoglobin recovery, transfusion requirements, and adverse events. Obstetric data collected included mode of delivery, gestational age, induction of labor, preeclampsia, previous cesarean delivery, and other risk factors for obstetric hemorrhage.

Exclusion criteria included having a bleeding disorder other than VWD or known thrombophilia; previous identification of the D1472H sequence variant on genetic testing; a history of thrombosis, cardiac disease, untreated hypertension, renal disease, or cerebrovascular accident; a platelet count of <100 × 10⁹/L; or a past allergic reaction to VWF concentrates or TXA. Participants were also excluded if they had undergone surgery within 8 weeks of enrollment; used antiplatelet drugs (excluding low-dose aspirin), anticoagulants, or nonsteroidal anti-inflammatories; or received treatment with desmopressin (DDAVP), cryoprecipitate, whole blood, plasma, or plasma derivatives containing substantial quantities of VWF within 5 days of enrollment. All participants provided written informed consent before enrollment.

Participants were not screened for the D1472H polymorphism as part of the study protocol; however, individuals with a known D1472H variant were excluded during screening. Additional clinical testing relevant to VWD classification is described in “Results.”

After confirmation of eligibility, participants were randomly assigned 1:1, to receive rVWF at delivery and on postpartum days 1 and 2, with TXA within 3 hours of delivery, or rVWF alone. Randomization was performed using a permuted block design with random block sizes of 4 to 6. The allocation sequence was generated by an unmasked statistician before the enrollment of any participant and securely stored to ensure allocation concealment. Once eligibility was confirmed, clinical coordinators enrolled participants within 72 hours of screening and assigned them to their randomized study group via a secure electronic web portal. Blinding or masking was not implemented; all drugs were administered in an open-label manner.

After randomization, all 20 participants received rVWF (Vonvendi, Takeda Pharmaceuticals USA Lexington, MA) 80 IU/kg using prepregnancy weight by standard IV technique over 5 to 10 minutes before neuraxial anesthesia or during active labor, and again 24 and 48 hours postpartum. All participants received rVWF at delivery and postpartum days 1 and 2 per study protocol, regardless of third-trimester VWF levels. A dose of 80 IU/kg was selected based on concerns regarding increased plasma volume during pregnancy and the observation that VWF activity levels in unaffected women typically rise significantly at delivery. This higher dose was intended to approximate physiologic third-trimester VWF activity levels and optimize hemostatic protection.

For the 7 participants who underwent cesarean delivery, rVWF was administered before neuraxial anesthesia preoperatively. All 13 participants who had vaginal deliveries elected for epidural anesthesia and received rVWF before catheter placement. The time from epidural placement to delivery was <12 hours for all but 1 participant who required an additional dose of rVWF 24 hours after initial dosing due to prolonged labor. Participants randomized to the rVWF + TXA group received IV TXA (Cyclokapron, Pfizer Pharmaceuticals, San Diego, CA), 1 g (100 mg/mL) infused over 10 minutes at a rate of 1 mL/min, administered within 3 hours of delivery. All participants in this group received the TXA dose within 1 hour postpartum (median, 35 minutes [range, 9-59]). An additional 1-g rescue dose of TXA was available for ongoing bleeding persisting beyond 30 minutes, at the discretion of the obstetric team.

There were no planned interim analyses. Assigned study drug was prepared for each participant, with rVWF stored at the Hemophilia Center of Western of Pennsylvania at 2 to 8°C and delivered to the investigational pharmacy at Magee-Womens Hospital for each participant before delivery. TXA was prepared and provided by the same investigational pharmacy.

Participants completed an initial screening visit and additional study visits on admission for delivery, postpartum days 1 and 2, and a final visit at 21 days postpartum. Laboratory assessments at each visit included hemoglobin, platelets, ferritin, D-dimer, fibrinogen, von Willebrand factor antigen (VWF:Ag), and VWF:RCo. Concomitant medications and narrative descriptions of postpartum management by the obstetrical team were collected. Quantitative blood loss (QBL) was determined by clinical staff as the sum of the blood volume collected in suction canisters (after subtracting irrigation fluids) and the weight of blood-soaked materials (subtracting standardized dry weight), and was recorded continuously from the time of delivery through at least 4 hours postpartum, in accordance with institutional protocols.³¹ Postpartum lochial bleeding was quantified by each participant, using a 21-day diary and PBAC.³² To standardize measurement of lochial blood loss, all participants received a 21-day supply of Kotex superplus sanitary pads (Kimberly-Clark, Irving, TX) for exclusive use during the trial.

Protocol amendments included expansion of inclusion criteria to include participants on low-dose aspirin therapy for obstetrical indication (ie, preeclampsia risk) and participant screening and enrollment beyond 32 weeks gestation. All participants receiving low-dose aspirin had this medication discontinued by their treated obstetrician between 35 and 37 weeks gestation and none delivered on active aspirin therapy. Stopping rules that, if reached, would halt the trial included thrombosis or grade 2 to 5 allergic reactions. Safety was assessed by lochial blood loss assessed by each participant by PBAC, blood product transfusion, thromboembolic event, allergic reaction, and hysterectomy within 21 days of delivery. Adverse events were graded by Common Terminology Criteria for Adverse Event, National Cancer Institute, grading, version 6.

Coagulation testing followed manufacturer protocols using a BCS-XP automated coagulation analyzer (Siemens). D-dimer was assessed via immunoturbidimetric assay (INNOVANCE D-Dimer kit, Siemens Healthineers, Malvern, PA), measuring change in optical density against calibration curves and reported in milligram per liter fibrinogen equivalent units. Fibrinogen activity was measured by the modified Clauss method (Multifibren U, Siemens Healthineers), in which the clotting time was compared with a fibrinogen standard and reported in milligram per deciliter. VWF:Ag and VWF:RCo were evaluated by agglutination assays (STA-Liatest VWF, BC von Willebrand Reagent, Siemens Healthcare Diagnostics, Marburg, Germany), with results expressed in units per milliliter. Activated partial thromboplastin time and FVIII activity were determined by clotting assays, using Pathromtin SL reagent (Siemens Healthineers) and FVIII-deficient plasma, calibrated by standard plasma curve.

The primary end point was QBL at delivery, calculated by a labor suite nurse using standardized methods.³³ Secondary end points included safety assessments, such as lochial blood loss quantified by PBAC, blood product usage, transfusion requirement, and hysterectomy within 21 days postpartum. Additionally, mechanisms of PPH reduction were explored by analysis of VWF:Ag, VWF:RCo, fibrinogen levels, and D-dimer values.

The statistical analysis for this pilot study, designed to assess the safety and feasibility of rVWF in combination with TXA at delivery in women with VWD, was exploratory and not powered for statistical significance. QBL, the primary end point, was analyzed using descriptive statistics and compared between treatment arms with a Mann-Whitney U test for nonnormal data. Secondary end points included 21-day lochial blood loss (PBAC) and safety outcomes (hemoglobin change, blood product usage, and hysterectomy), analyzed using descriptive statistics and appropriate parametric or nonparametric tests. Stratified analyses assessed QBL by age, body mas index (BMI) at delivery, baseline VWF activity, and delivery type. Analyses were performed in STATA (version 16, StataCorp, College Station, TX), with results focused on trends. Additional methods and data are available in the online supplemental material.

During the preparation of this work, the authors used ChatGPT for editing and text refinement. After using this tool, the author(s) reviewed and edited the content as needed and take full responsibility for the content of the publication.

Between 4 June 2021 and 17 May 2024, 103 women were screened; 40 met eligibility criteria; and 20 were randomized, 10 to receive rVWF with TXA, and 10 to receive rVWF alone (Figure 1). All participants completed the study protocol and attended all study visits. The most common reasons for screen failure included delivery at an outside facility and inability to verify a VWD diagnosis from historical laboratory results. Of 40 eligible women, 20 declined participation, most often due to concerns about added demands during pregnancy. This was more commonly expressed by women in their first pregnancy, based on consent discussions. Full screening outcomes are shown in Figure 1.

Two protocol deviations occurred: 1 participant in the rVWF + TXA group received an additional 1-g dose of IV TXA during epidural placement, in addition to the study dose administered within 3 hours of delivery. Another participant in the rVWF-alone group received a rescue dose of 80 IU/kg rVWF after >24 hours of labor. These participants were included in the analysis.

QBL data were available for all participants (20/20 [100%]), and 21-day PBAC data were available for 19 of 20 participants (95%). Of 200 scheduled laboratory draws across 20 participants, only 4 values (2%) were missing.

Baseline demographic and clinical characteristics were comparable between the 2 treatment groups (Table 1). The mean age of participants was 29.0 ± 5.23 years (± standard deviation), with no significant difference between the rVWF + TXA group (29.1 ± 5.2 years,) and the rVWF-alone group (28.8 ± 5.7 years). Most participants (90%) were White, and all participants had type 1 VWD.

Ferritin levels obtained in the third trimester indicated universal iron deficiency among participants, with a mean ferritin level of 15.8 ± 9.1 ng/mL. Although all participants reported taking a daily prenatal vitamin, only 2 were on formulations containing iron. None had undergone ferritin assessment in the previous year, were prescribed iron supplementation, or received recommendations for iron replacement before enrollment. Ferritin levels were <50 ng/mL in all 20 participants; of these, 19 (95%) received IV iron during the third trimester before delivery. One participant declined IV iron and was treated with oral ferrous sulfate instead.

The mean prepregnancy BMI was 30.9 ± 8.4, increasing to 36.1 ± 8.1 by delivery, with no significant difference between groups. Third-trimester hemoglobin levels were also comparable (12.1 ± 0.9 g/dL in the rVWF + TXA group vs 11.8 ± 1.0 g/dL in the rVWF-alone group).

Among 20 enrolled participants, 11 had previous clinical next-generation sequencing confirming absence of the D1472H variant. An additional 3 participants, although lacking next-generation sequencing data, had von Willebrand factor GP1bM (VWF:GP1bM) activity of <0.5 IU/mL outside of pregnancy, reducing concern for misclassification due to this polymorphism. Thus, 14 of 20 participants had available clinical data supporting accurate VWD classification (Table 2).

QBL at delivery was numerically higher in the rVWF + TXA group compared with the rVWF-alone group, with a median QBL of 802.5 mL (range, 350.0-1027.5) vs 355.1 mL (range, 255.0-499.3), respectively (P = 0.41); this difference was not statistically significant. PPH, defined as QBL of >1000 mL, occurred in 3 participants (30%) in the rVWF + TXA group and 1 participant (10%) in the rVWF-alone group (P = 0.25) (Table 3).

Obstetricians initiated additional interventions for bleeding in 6 participants, with 3 participants (30%) in each group (Table 2). All received uterotonics, including misoprostol and oxytocin. Intrauterine tamponade balloon placement was required in 2 participants. In the rVWF + TXA group, 1 participant received a rescue dose of TXA for PPH within 24 hours after delivery. In the rVWF-alone group, 3 participants received 1 g of IV TXA within 3 hours of delivery due to clinical concern for primary PPH, per standard hospital protocol; all were included in the final analysis. No participant required additional doses of rVWF outside of the study protocol.

Cesarean delivery was significantly more common in the rVWF + TXA group (60%) compared with the rVWF-alone group (10%; P = 0.027), with vaginal deliveries accounting for the remaining 40% and 90% of births, respectively. Stratification of QBL by delivery type revealed higher mean QBL among participants undergoing cesarean delivery (903.0 ± 285.0 mL [SD]) compared with those with vaginal delivery (462.5 ± 459.0 mL [SD]; P = 0.13). Stratified analyses of QBL by delivery type, BMI, and age revealed no significant subgroup differences.

Hemoglobin levels declined by a mean of 1.9 ± 1.5 g/dL[SD]in the rVWF + TXA group and 1.4 ± 1.6g/dL[SD] in the rVWF-alone group; this difference was not statistically significant (P = 0.49). The proportion of participants experiencing a >2 g/dL hemoglobin decrease was slightly higher in the rVWF + TXA group (40%) than in the rVWF-alone group (30%; P = 0.32). The median cumulative 21-day PBAC score was 442.0 (range, 224.0-724.0) in the rVWF + TXA group and 344.8 (range, 185.0-512.5) in the rVWF-alone group (P = 0.32). The duration of lochial bleeding was similar between groups: 19.1 ± 3.1 days in the rVWF + TXA group and 19.0 ± 5.4 days [SD] (P = 0.96). The mean PBAC scores over time were comparable between groups (Figure 2). On day 21, 8 women had ongoing lochial bleeding: 4 in the rVWF + TXA group, and 5 in the rVWF-alone group. One participant in the rVWF + TXA group did not complete the PBAC diary and was excluded from this analysis.

Laboratory markers of coagulation, including VWF:Ag, VWF:RCo, FVIII, fibrinogen, and D-dimer, did not differ significantly between groups across study visits (Table 4).

Participants were divided into those clinically managed for PPH (6) and those who were not (14) to explore potential differences in characteristics and coagulation testing. The group treated for PPH by the obstetrician included all 4 participants with QBLs of >1000 mL, whereas the other 2 participants had QBLs of 800 mL and 300 mL, respectively. Within the group managed for PPH, 3 participants had cesarean deliveries and 3 had vaginal deliveries. The age (26.6 ± 5.2 years [SD] vs 30.0 ± 5.1 years [SD]; P = 0.15) and BMI (37.9 ± 12.0 kg/m² [SD] vs 35.3 ± 6.3 kg/m² [SD]; P = 0.63) of the participants with and without PPH were similar. There was no difference in third trimester VWF:Ag, VWF:RCo, FVIII, D-dimer, or fibrinogen levels between participants who were and were not managed clinically for PPH. Evaluation VWF:Ag over study times points did demonstrate lower VWF:Ag (1.39 ± 0.40 [SD]) on admission for delivery for participants with PPH than those who did not (2.45 ± 1.07 [SD]), P = 0.006. There was no difference between groups for fibrinogen, D-dimer, VWF:RCo, FVIII, or VWF:Ag across other study time points.

No serious adverse events, thrombotic events, or allergic reactions were observed. No hysterectomies occurred within the 21-day postpartum period. Of 19 participants who received IV iron there were no infusion reactions or complications.

Despite current prophylactic approaches, women with VWD remain at high risk for bleeding at delivery. In this pilot randomized trial, we evaluated the feasibility and impact of combining TXA with rVWF at delivery. Substantial blood loss persisted in both treatment groups, with a median QBL exceeding 700 mL in the combination arm, and one-third of all participants requiring additional clinical intervention for bleeding. These findings highlight the limitations of current therapies in fully preventing postpartum bleeding in VWD. As this study enrolled only women with type 1 VWD, the applicability of these findings to more severe subtypes remains uncertain.

The high proportion of these VWD participants with iron deficiency in the third trimester of pregnancy was notable and aligns with growing awareness of iron deficiency among pregnant women.³⁴ However, the absence of guidelines by the United States Public Health Service Commissioned Corps and controversy over standardization of screening and management appears to contribute to the lack of iron screening and repletion as women with VWD approach delivery.³⁵ Until guidelines are available, routine screening should be implemented during the second and third trimester, with repletion as necessary, given the poor quality of life and morbidity associated with iron deficiency and growing evidence of impact on fetal outcomes.³⁶ 

There are several limitations of this study. First, the small sample size and pilot nature of this study restrict the generalizability of the findings. Subgroup and stratified analyses were underpowered and should be interpreted as exploratory. Larger, multicenter studies will be necessary to validate these results. Furthermore, because findings in type 1 VWD may not predict those with more severe disease, future studies should include pregnant females with type 2 and 3 VWD. The high rate of cesarean deliveries in the TXA group may have influenced the QBL results, introducing a potential confounder that should also be addressed in larger future trials. Participants receiving VWF alone who subsequently developed PPH were managed with TXA, which could influence coagulation assay testing and PBAC scores postpartum. Moreover, the lack of understanding of the mechanism of PPH in VWD and the lack of objective predictors of PPH highlight the complexity of designing a trial to optimize PPH management in women with VWD. Despite VWF and TXA coadministration, the PPH rate remains high at 20%, suggesting a need for better therapies, better dosing strategies, and better pharmacologic and procedural approaches to prevent PPH in VWD.

Consistent with current clinical guidelines and previous studies, all participants in our trial were treated with a regimen designed to achieve VWF and FVIII levels of >50% to permit safe neuraxial anesthesia and provide hemostatic protection.^15,16 Although our study was not designed to compare different factor level targets, the persistent rate of PPH despite achieving this threshold suggests that a target of >50% may be insufficient for some individuals with VWD. Recent studies have proposed aiming for higher factor levels to better mimic physiologic third-trimester hemostasis, although the optimal target remains uncertain.^24,37 Emerging data from European cohorts and the recently reported Postpartum Hemorrhage in Women with Von Willebrand Disease after Enhanced Prophylactic Clotting Factor Suppletion: The Pregnancy and Inherited Bleeding Disorders Study (PRIDES) suggest that more intensive or repeated postpartum VWF replacement may better reflect physiologic hemostatic changes and could reduce PPH risk.³⁸ These evolving practices support the rationale for our protocol’s repeated rVWF dosing.

Recent data suggest that the antifibrinolytic efficacy of TXA may be reduced in states of low hematocrit, due to decreased reinforcement of clot structure by red blood cells. In our cohort, although hemoglobin levels were modestly reduced, 100% of participants were iron deficient, which may have influenced TXA response. Future studies should explore whether hematocrit optimization or TXA dosing adjustments in individuals who are iron deficient could improve bleeding outcomes.³⁹ 

Our protocol prioritized immediate postpartum bleeding risk, with TXA administered within hours of delivery to target peak fibrinolysis. However, a single inpatient dose is unlikely to influence bleeding patterns across the broader postpartum period. Future studies should evaluate extended TXA regimens to address delayed or secondary PPH.

Despite these considerations, the study demonstrated the feasibility of conducting a randomized trial in pregnant patients with VWD, with full participant retention, high compliance with study protocol visits, laboratory tests, treatment, and safety with combined therapy interventions. The trial greatly benefited from its conduct in a women’s hospital with a team of collaborative anesthesiologists, obstetricians, and pharmacists, suggesting this successful approach in future larger-scale studies.

A notable finding in this cohort was the 100% prevalence of iron deficiency, and, despite the absence of guidelines, vigilance is needed for screening and management for women with VWD, because it represents a low-cost, high-yield intervention to improve maternal outcome and reduce the burden of, and hasten recovery from, associated bleeding-related complications.^40,41 

In conclusion, although the study provides encouraging data on the safety of conducting clinical trials in pregnant women with VWD, as well as the safety of combining rVWF and TXA in postpartum management for women with type 1 VWD, the limitations underscore the necessity for larger, more diverse studies including other VWD subtypes, as well as studies designed to evaluate hematologic modifiers of TXA efficacy such as hematocrit and red cell mass, and to explore optimized dosing, novel therapeutics, and iron deficiency management strategies to reduce the unacceptably high PPH risk.

The authors are grateful to the women who participated in this trial, whose involvement was essential to the success of the study. The authors are appreciative of the staff and clinicians at University of Pittsburgh Medical Center Magee-Womens Hospital for their invaluable assistance during labor, delivery, and the postpartum period. Their collaboration and dedication were critical in facilitating this research.

Funding for this investigator-initiated trial was by Takeda Pharmaceuticals, who also provided recombinant von Willebrand factor (Vonvendi) for the study.

The funding source did not have a role in the study design, execution, data collection, analysis, or interpretation of the data. However, the funding source reviewed and provided comments on the manuscript. The authors had full access to all data in the study and had final responsibility for the decision to submit for publication.

Contribution: All authors contributed to the conceptualization, methodology, formal analysis, and interpretation of results; N.C.M., M.M.B., and M.V.R directly accessed and verified the underlying data reported in the manuscript; all authors confirm that they had full access to all the data in the study and accept responsibility for submitting the manuscript for publication; N.C.M. led data collection, statistical analysis, and manuscript drafting (writing of original draft). D.V., D.I., and B.L. contributed to data collection and management; M.M.B. provided expertise in statistical methodology and data analysis (formal analysis); C.D.S. and F.X. provided clinical oversight, hematologic interpretation, and critical manuscript review (review and editing); S.S. and A.V. contributed expertise in laboratory and mechanistic aspects of the study in addition to critical manuscript review (review and editing); M.V.R. supervised the study, provided critical revisions (review and editing), and secured funding acquisition; and all authors reviewed and approved the final manuscript.

Conflict-of-interest disclosure: N.C.M. has received research funding from Takeda Pharmaceuticals USA; and has served on advisory boards for Sanofi and Star Therapeutics. M.V.R. has received research funding from Takeda Pharmaceuticals USA. F.X. has received consulting fees from Genentech and Sanofi. C.D.S. has received consulting fees from Takeda, CSL Behring, and Pfizer. The remaining authors declare no competing financial interests.

Correspondence: Nicoletta C. Machin, Hemophilia Center of Western PA, Sterling Plaza, 201 N Craig St Suite 500, Pittsburgh, PA 15213; email: machinnc2@upmc.edu.