CC BY 4.0 · Thromb Haemost 2022; 122(06): 913-925
DOI: 10.1055/s-0041-1739514
Coagulation and Fibrinolysis

Predictive Modeling Identifies Total Bleeds at 12-Weeks Postswitch to N8-GP Prophylaxis as a Predictor of Treatment Response

1   Katharine Dormandy Haemophilia and Thrombosis Centre, Royal Free Hospital, London, United Kingdom
Kingsley Hampton
2   Department of Cardiovascular Science, University of Sheffield, Sheffield, United Kingdom
Victor Jiménez-Yuste
3   Department of Hematology, La Paz University Hospital-IdiPaz, Autónoma University, Madrid, Spain
Guy Young
4   Hemostasis and Thrombosis Center, Cancer and Blood Disorders Institute, Children's Hospital Los Angeles, University of Southern California Keck School of Medicine, Los Angeles, California, United Sates
Soraya Benchikh el Fegoun
5   Global Medical Affairs Biopharm, Novo Nordisk Health Care AG, Zürich, Switzerland
Aidan Cooper
6   Predictive Analytics, Real World Solutions, IQVIA, London, United Kingdom
Erik Scalfaro
7   Real World Insights, IQVIA, Basel, Switzerland
8   Department of Hematology, Hemostasis, Oncology and Stem Cell Transplantation, Hannover Medical School, Hanover, Germany
› Author Affiliations
Funding This analysis and medical writing support for the article was funded by Novo Nordisk A/S (Bagsværd, Denmark).


Background Predicting annualized bleeding rate (ABR) during factor VIII (FVIII) prophylaxis for severe hemophilia A (SHA) is important for long-term outcomes. This study used supervised machine learning-based predictive modeling to identify predictors of long-term ABR during prophylaxis with an extended half-life FVIII.

Methods Data were from 166 SHA patients who received N8-GP prophylaxis (50 IU/kg every 4 days) in the pathfinder 2 study. Predictive models were developed to identify variables associated with an ABR of ≤1 versus >1 during the trial's main phase (median follow-up of 469 days). Model performance was assessed using area under the receiver operator characteristic curve (AUROC). Pre-N8-GP prophylaxis models learned from data collected at baseline; post-N8-GP prophylaxis models learned from data collected up to 12-weeks postswitch to N8-GP, and predicted ABR at the end of the outcome period (final year of treatment in the main phase).

Results The predictive model using baseline variables had moderate performance (AUROC = 0.64) for predicting observed ABR. The most performant model used data collected at 12-weeks postswitch (AUROC = 0.79) with cumulative bleed count up to 12 weeks as the most informative variable, followed by baseline von Willebrand factor and mean FVIII at 30 minutes postdose. Univariate cumulative bleed count at 12 weeks performed equally well to the 12-weeks postswitch model (AUROC = 0.75). Pharmacokinetic measures were indicative, but not essential, to predict ABR.

Conclusion Cumulative bleed count up to 12-weeks postswitch was as informative as the 12-week post-switch predictive model for predicting long-term ABR, supporting alterations in prophylaxis based on treatment response.

Author Contributions

All authors made substantial contributions to the conception and design, acquisition of data, or analysis and interpretation of data; took part in drafting the article or revising it critically for important intellectual content; gave final approval of the version to be published; and agree to be accountable for all aspects of the work. The underlying data included in the analyses have been verified by A. Cooper and E. Scalfaro.


Novo Nordisk is the proprietary owner of all raw data analyzed. Transformed data used for machine learning model development are available from the corresponding author, P.C., upon reasonable request and subject to approval by Novo Nordisk. Underlying data from the pathfinder 2 trial, which formed the data source for this analysis, will be shared with bona fide researchers submitting a research proposal requesting access to data. The access request proposal form and the access criteria can be found at Data will be available permanently after research completion and approval of product and product use in both the European Union and United States on a specialized Statistical Analysis System data platform. The analyses available for use will be those as approved by the Independent Review Board according to the IRB Charter (see Individual participant data will be shared in datasets in a de-identified/-anonymized format. In addition, the study protocol and redacted Clinical Study Report will be available according to Novo Nordisk data sharing commitments.

Supplementary Material

Publication History

Received: 16 April 2021

Accepted: 03 October 2021

Article published online:
05 December 2021

© 2021. The Author(s). This is an open access article published by Thieme under the terms of the Creative Commons Attribution License, permitting unrestricted use, distribution, and reproduction so long as the original work is properly cited. (

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany