Background: Von Willebrand disease (VWD) is the most common inherited bleeding disorder and is characterized by deficient and/or defective von Willebrand factor (VWF) which results in spontaneous and traumatic mucosal bleeding. In women with VWD, pregnancy is associated with excess blood loss and poor quality of life. Recently, a prospective cohort study by James et al Haemophilia 2015, determined that even when treatment is given, women with VWD had lower VWF levels, greater blood loss at delivery, and lower hematocrit than controls without VWD. The reason for this finding remains unknown. Current VWF dosing is weight-based, but does not account for the ~1.4-1.5-fold increase in blood volume during pregnancy. To address this, we conducted a feasibility study for a prospective, randomized phase III trial comparing weight-based, 50 IU/kg, with volume-based, 80 IU/kg, VWF to prevent postpartum hemorrhage (PREVENT PPH Trial).
Methods: To establish trial feasibility: 1) we compared pre-pregnancy and 8th-month VWF levels in women with VWD with and without PPH (≥500 cc blood loss in the 1st 24 hours) following vaginal delivery; 2) we evaluated VWF dosing in women with PPH in the James study; 3) we assessed thrombosis risk and VWF concentrate dose by literature review; 4) we surveyed current practice regarding VWF concentrate use at delivery by U.S. hemophilia treatment center (HTC) MDs; and 5) we conducted structured interviews of MDs and VWD patients to determine trial acceptability. Analysis was by Student's t-test for continuous data, and Fisher's exact test for discrete data.
Results:In the Retrospective VWD-PPH Study of 16 women with VWD (14 type 1; 2 type 2) undergoing vaginal delivery, PPH was associated with higher pre-delivery weight, pre: 88.1 vs. 67.6 kg; 8thmonth: 99.9 vs. 75.0 kg; and delivery: 104.0 vs. 78.6 kg, all p<0.005; and a family bleeding history, 75.0% vs. 12.5%, p=0.041. Women with PPH had lower pre-pregnancy VWF:RCo, 0.34 IU/ml vs. 0.48 IU/ml, non-significant, p=0.067; similar 8th month VWF:RCo, 1.31 IU/ml vs. 1.45 IU/ml, p=0.484, and were more likely to be treated with VWF, 75.0% vs. 37.5%, p=0.315. Pre-pregnancy bleeding score (BS) correlated directly with EBL (blood loss) at delivery, r=0.663, p=0.005. VWF Dosing Data from James co-authors indicated a mean VWF dose of 46 IU/kg (median 50 IU/kg), range 28-83 IU/kg, at delivery, followed by 3.8 days (median 3 days), range 2-19 days, postpartum treatment. The Literature Review identified 13 publications between 1992-2015 reporting pharmacokinetic, safety, and/or efficacy studies in a total of 570 patients (none pregnant) at VWF doses 10-222 IU/kg. Thrombosis rate was low, 0.4%, including 2 of 353 (0.6%) receiving plasma-derived (pd) VWF and none of 95 (0%) receiving recombinant (r) VWF: these two patients included one with injection-site phlebitis and one with a remote postoperative VTE, each receiving VWF≤100 IU/kg. An HTC Survey distributed to 70 MDs, 19 (27.1%) of whom responded, 16/18 (88.9%) reported VWF was first-line therapy at delivery, mean dose 50 IU/kg; DDAVP was the most common second-line therapy in 7/17 (41.2%), and anti-fibrinolytic therapy was third-line, 8/10 (80.0%). All 19 indicated interest and willingness to participate in a trial comparing weight-based, 50 IU/kg, vs. 1.6-fold higher volume-based, 80 IU/kg, VWF at delivery. In Structured Interviews of 18 MDs and 18 VWD patients, the trial was acceptable to MDs if staff costs are covered, if the drug is safe re thrombosis, if there are sufficient patients, and if there is obstetrician collaboration to facilitate in-hospital dosing and blood draws. The trial was also acceptable to VWD patients if the drug is safe for mother and child, if childcare and travel costs are covered, and if visits are minimized in the postpartum period.
Conclusions: The findings of this feasibility study indicate that pre-pregnancy VWF:RCo may be a better predictor of PPH than 8thmonth VWF:RCo. High pre-delivery weight which is associated with high blood volume may increase PPH, possibly through dilution of VWF levels. A VWF dose of ~50 IU/kg, which is the first line of therapy at delivery and the current standard of practice, may not prevent PPH. Thrombosis risk with VWF in published studies is low even at doses >80 IU/kg. Drug safety, local obstetric collaboration, and coverage of staff, travel, and childcare costs are critical issues to address in trial design.
Ragni: SPARK: Research Funding; Genentech: Research Funding; Tacere Benitec: Consultancy; Vascular Medicine Institute: Research Funding; OPKO: Research Funding; Novo Nordisk: Research Funding; Alnylam Pharmaceuticals: Consultancy, Research Funding; Baxalta: Research Funding; Biogen: Consultancy, Research Funding; Biomarin: Consultancy; Shire: Consultancy; CSL Behring: Research Funding. James:Intramural University of Ghana Research Fund: Research Funding; Vanderbilt University Medical Center Gift Funds: Research Funding. Malec:Biogen: Consultancy; Vascular Medicine Institute: Research Funding; Biogen: Research Funding; Baxalta: Research Funding. Kessler:Octapharma: Consultancy, Research Funding; Novo Nordisk: Consultancy, Research Funding; Grifols: Consultancy; Genentech: Consultancy, Research Funding; Biogen: Consultancy; Bayer: Consultancy, Research Funding; Pfizer: Consultancy; Baxalta: Consultancy, Research Funding; LFB: Other: Member of DSMB. Kouides:CSL Behring: Honoraria, Membership on an entity's Board of Directors or advisory committees; Baxalta: Consultancy. Neff:Shire: Membership on an entity's Board of Directors or advisory committees; Pfizer: Other: DSMB Chair for research study; ABIM: Other: Hematology Exam committee; CSL Behring: Membership on an entity's Board of Directors or advisory committees; HEMA Biologics: Membership on an entity's Board of Directors or advisory committees. Philipp:Novo Nordisk: Membership on an entity's Board of Directors or advisory committees; Baxalta: Membership on an entity's Board of Directors or advisory committees, Research Funding; Merck: Other: Data Safety Monitoring Board. Brooks:Gilead Sciences: Research Funding.