How Can the Threshold Method Guide the Choice of Treatments? Illustration in French Patients With Hemophilia A or B
Author(s)
Yasmine Fahfouhi, MSc1, Anais Reynaud, PhD1, Astrid Foix-Colonier, MSc2, Alaeddine Sidhom3, Jérémie Rudant, PhD, MD1, Nicolas Giraud, MSc4, Benoit Guillet, MD5, Aline Gauthier, MSc6, Monia Ezzalfani, PhD7, Yesim Dargaud, MD, PhD8.
1Pfizer, Paris, France, 2Amaris consulting, Saint-Herblain, France, 3Barcelona, Spain, 4Association française des hémophiles, Paris, France, 5Centre de référence hémophilie et autres déficits constitutionnels en protéines de la coagulation, Hospices Civils de Lyon, France, 6Amaris Consulting, Barcelona, Spain, 7Amaris, Paris, France, 8Centre de référence de l'hémophilie et des maladies hémorragiques, CHU Rennes, France.
1Pfizer, Paris, France, 2Amaris consulting, Saint-Herblain, France, 3Barcelona, Spain, 4Association française des hémophiles, Paris, France, 5Centre de référence hémophilie et autres déficits constitutionnels en protéines de la coagulation, Hospices Civils de Lyon, France, 6Amaris Consulting, Barcelona, Spain, 7Amaris, Paris, France, 8Centre de référence de l'hémophilie et des maladies hémorragiques, CHU Rennes, France.
OBJECTIVES: Preference studies are essential tools for supporting public decision-making, for optimal allocation of resources. This work illustrates one of the most recognized methods: the Threshold Technique (TT), using an example in hemophilia.
METHODS: The TT is based on the identification of treatment characteristics (attributes) and their levels (e.g., annual bleeding rate [ABR] for the attribute, and 5 bleeds/year for the corresponding level). When choosing, patients compare a fixed reference treatment to a target treatment in which only one attribute varies. TT allows to estimate, through interval regression, the minimum acceptable benefit or the maximum acceptable risk (MAR) for a clinical benefit or risk, respectively, as well as their confidence intervals (CI). Results can be expressed as a percentage of patients per threshold interval and as target treatment acceptance rates based on attribute levels. A recent application to 86 French patients with hemophilia A or B evaluated the preference thresholds for each attribute associated with the different treatments: replacement factor (FRT), bispecific antibody (BA), gene therapy, new conventional therapies (NT) including subcutaneous (SC) and intravenous.
RESULTS: The attributes related to the reference treatments BA and FRT were set according to Table 1. For the NT SC target treatment, all attributes were fixed except for the ABR, which was varied to estimate MAR. When switching from BA or FRT to NT SC, 26.9% and 58.1% of patients respectively would accept NT SC with 2 bleeds/year. The mean MAR was 1.5 (95% CI: 1-2) bleeding/year with a median of 0.5 when switching from BA to NT SC, and 3.3 (95% CI: 2.7-4) bleeding/year with a median of 3.5 when switching from FRT to NT SC.
CONCLUSIONS: Studying patient preferences is a robust approach to integrate the patients’ voice into therapeutic decision-making, especially for complex diseases for which the therapeutic arsenal is important, such as hemophilia.
METHODS: The TT is based on the identification of treatment characteristics (attributes) and their levels (e.g., annual bleeding rate [ABR] for the attribute, and 5 bleeds/year for the corresponding level). When choosing, patients compare a fixed reference treatment to a target treatment in which only one attribute varies. TT allows to estimate, through interval regression, the minimum acceptable benefit or the maximum acceptable risk (MAR) for a clinical benefit or risk, respectively, as well as their confidence intervals (CI). Results can be expressed as a percentage of patients per threshold interval and as target treatment acceptance rates based on attribute levels. A recent application to 86 French patients with hemophilia A or B evaluated the preference thresholds for each attribute associated with the different treatments: replacement factor (FRT), bispecific antibody (BA), gene therapy, new conventional therapies (NT) including subcutaneous (SC) and intravenous.
RESULTS: The attributes related to the reference treatments BA and FRT were set according to Table 1. For the NT SC target treatment, all attributes were fixed except for the ABR, which was varied to estimate MAR. When switching from BA or FRT to NT SC, 26.9% and 58.1% of patients respectively would accept NT SC with 2 bleeds/year. The mean MAR was 1.5 (95% CI: 1-2) bleeding/year with a median of 0.5 when switching from BA to NT SC, and 3.3 (95% CI: 2.7-4) bleeding/year with a median of 3.5 when switching from FRT to NT SC.
CONCLUSIONS: Studying patient preferences is a robust approach to integrate the patients’ voice into therapeutic decision-making, especially for complex diseases for which the therapeutic arsenal is important, such as hemophilia.
Conference/Value in Health Info
2025-11, ISPOR Europe 2025, Glasgow, Scotland
Value in Health, Volume 28, Issue S2
Code
MSR119
Topic
Clinical Outcomes, Methodological & Statistical Research, Patient-Centered Research
Disease
Rare & Orphan Diseases