A METHODOLOGY FOR DEVELOPING COMPOSITE ENDPOINT SELECTION CRITERIA INTEGRATING MULTI-STAKEHOLDER PERSPECTIVES
Author(s)
Eleanor Scouler, MSc1, Adeebah Adams, MD2, Daphne Robakis, MD3, Duska Franic, PharmD, MS, PhD3, Kathleen Hannon, PhD1, Chad Gwaltney, PhD4, Olivier Chassany, PhD, MD5, Rob Arbuckle, MA, MSc1, Benoit Arnould, PhD6.
1Adelphi Values Patient-Centered Outcomes, Macclesfield, Cheshire, United Kingdom, 2Sanofi, Riyad, Saudi Arabia, 3Sanofi, Morristown, NJ, USA, 4Gwaltney Consulting, Westerly, RI, USA, 5Patient-Reported Outcomes Research (PROQOL), Health Economics Clinical Trial Unit (URC-ECO), Hotel-Dieu Hospital, AP-HP, Université Paris-Cité, Paris, France, 6Sanofi, Lyon, France.
1Adelphi Values Patient-Centered Outcomes, Macclesfield, Cheshire, United Kingdom, 2Sanofi, Riyad, Saudi Arabia, 3Sanofi, Morristown, NJ, USA, 4Gwaltney Consulting, Westerly, RI, USA, 5Patient-Reported Outcomes Research (PROQOL), Health Economics Clinical Trial Unit (URC-ECO), Hotel-Dieu Hospital, AP-HP, Université Paris-Cité, Paris, France, 6Sanofi, Lyon, France.
OBJECTIVES: Composite endpoints combine multiple distinct outcome measures (called components) into a single endpoint, enabling comprehensive assessment of treatment benefit in clinical trials. Component selection and aggregation is commonly based on clinical consensus. Despite widespread use, no established selection methods exist to secure fit-for-purpose composite endpoints. Systematic application of pre-defined criteria, based on relevant multi-stakeholder input, will ensure the selection of the most appropriate, clinically relevant, patient-centered component measures. This research aimed to develop a methodology for developing composite endpoint selection criteria integrating perspectives from measurement experts, clinicians, and patients.
METHODS: A targeted literature review (TLR) identified existing endpoint selection considerations. Draft criteria were then iteratively refined through four stakeholder activities: two measurement expert panels (n=3), one patient representative panel (n=2 patients, n=1 care partner), and one clinical expert panel (n=3). All panels were conducted via video call with pre-read materials. Structured feedback was obtained on inclusion/exclusion of each draft criterion, wording refinements, and identification of missing criteria. Post-panel ranking exercises prioritized criteria by importance.
RESULTS: The TLR identified no formal criteria but yielded 10 relevant publications offering insights into composite endpoint development. 15 draft criteria were developed, refined to 12 through measurement expert consultation. Patient and clinician input led to wording refinements of the 12 criteria, with all criteria considered relevant by both groups of stakeholders. Wording was further refined during a final measurement expert panel. The final 12 criteria address: patient relevance and clinical meaningfulness (n=2); sensitivity to detect change (n=1); measurement feasibility and accessibility (n=1); psychometric robustness (n=2); interpretability (n=1); component weighting (n=3); disease prevalence (n=1); and ability to detect treatment effects across early efficacy/registrational trials (n=1).
CONCLUSIONS: This multi-stakeholder approach provides the first methodology for establishing composite endpoint selection criteria integrating priorities of expert clinicians, measurement specialists, and most importantly patients - ensuring patient-centred, clinically meaningful endpoints.
METHODS: A targeted literature review (TLR) identified existing endpoint selection considerations. Draft criteria were then iteratively refined through four stakeholder activities: two measurement expert panels (n=3), one patient representative panel (n=2 patients, n=1 care partner), and one clinical expert panel (n=3). All panels were conducted via video call with pre-read materials. Structured feedback was obtained on inclusion/exclusion of each draft criterion, wording refinements, and identification of missing criteria. Post-panel ranking exercises prioritized criteria by importance.
RESULTS: The TLR identified no formal criteria but yielded 10 relevant publications offering insights into composite endpoint development. 15 draft criteria were developed, refined to 12 through measurement expert consultation. Patient and clinician input led to wording refinements of the 12 criteria, with all criteria considered relevant by both groups of stakeholders. Wording was further refined during a final measurement expert panel. The final 12 criteria address: patient relevance and clinical meaningfulness (n=2); sensitivity to detect change (n=1); measurement feasibility and accessibility (n=1); psychometric robustness (n=2); interpretability (n=1); component weighting (n=3); disease prevalence (n=1); and ability to detect treatment effects across early efficacy/registrational trials (n=1).
CONCLUSIONS: This multi-stakeholder approach provides the first methodology for establishing composite endpoint selection criteria integrating priorities of expert clinicians, measurement specialists, and most importantly patients - ensuring patient-centred, clinically meaningful endpoints.
Conference/Value in Health Info
2026-11, ISPOR Europe 2026, Vienna, Austria
Value in Health, Volume 29, Issue 12S
Code
CO140
Topic
Clinical Outcomes, Patient-Centered Research, Study Approaches
Topic Subcategory
Clinical Outcomes Assessment
Disease
No Additional Disease & Conditions/Specialized Treatment Areas