The Relationship Between Surrogate Endpoints and Overall Survival in Myelofibrosis Clinical Trials: Results From a Targeted Literature Review
Author(s)
Louise Linsell, DPhil, MSc, BSc1, Yuxian Chen, MSc1, Federico Felizzi, BSc, MS, PhD2, Santiago Moreno, PhD3, Elisabetta Fenu, MSc4, MARIA PAPAGEORGIOU, MSc, MBA5, Dave Ray, PharmD, MBA, MPH6, Neil Hawkins, MBA, MSc, PhD7, Keith R. Abrams, BSc, MSc, PhD8.
1Visible Analytics Ltd, Oxford, United Kingdom, 2Global HEOR Franchise Lead - Oncology, Menarini, Zurich, Switzerland, 3Menarini, Zurich, Switzerland, 4Menarini Group, Zürich, Switzerland, 5Menarini Stemline, MAROUSI, Greece, 6Karyopharm Therapeutics, Newton, MA, USA, 7University of Glasgow, Oxford, United Kingdom, 8University of Warwick, Coventry, United Kingdom.
1Visible Analytics Ltd, Oxford, United Kingdom, 2Global HEOR Franchise Lead - Oncology, Menarini, Zurich, Switzerland, 3Menarini, Zurich, Switzerland, 4Menarini Group, Zürich, Switzerland, 5Menarini Stemline, MAROUSI, Greece, 6Karyopharm Therapeutics, Newton, MA, USA, 7University of Glasgow, Oxford, United Kingdom, 8University of Warwick, Coventry, United Kingdom.
OBJECTIVES: : In myelofibrosis, overall survival (OS) is the gold-standard endpoint; however, its delayed manifestation challenges timely evaluation of treatment benefit. This scoping literature review aimed to explore the potential of intermediate clinical outcomes—including SVR35 (≥35% spleen volume reduction), symptom response, change in bone marrow fibrosis (BMF), transfusion independence, and anemia response—as surrogate endpoints for OS.
METHODS: A targeted literature review employed multiple search strategies: EMBASE searches for surrogate endpoints and meta-analyses/network meta-analyses, supplemented by ChatGPT queries (with manual verification), Google Scholar searches, and review of myelofibrosis clinical guidelines and relevant NICE appraisals. Publications were screened for evidence on associations between post-treatment changes in intermediate clinical outcomes and overall survival.
RESULTS: A total of 11 publications (9 unique studies) were identified from EMBASE, with an additional 4 relevant studies found through ChatGPT-assisted searches. These included clinical trials and real-world studies highlighting associations between SVR (10% to 50% spleen volume reduction), BMF, transfusion status, anemia response, and OS. SVR was shown to be predictive of survival outcomes in several studies, both in JAK-inhibitor-naïve and -experienced populations. It was also identified as a key factor in multiple prognostic models and observational cohorts. Two network meta-analyses involving 9 randomized controlled trials incorporated studies reporting intermediate endpoints (SVR, anemia-related metrics) and OS outcomes, providing a potential evidence base for future surrogate endpoint validation analyses.
CONCLUSIONS: The review identifies potential surrogate endpoints - in particular SVR, and transfusion status - associated with predicting OS in patients with myelofibrosis. SVR emerges as a particularly relevant surrogate, given its strong association with disease burden and patient outcomes. These findings underscore the importance of these surrogate endpoints when considering the clinical benefit of treatment for patients with MF. Further validation using multivariate meta-analysis is recommended to quantify the strength of the surrogate relationship with OS.
METHODS: A targeted literature review employed multiple search strategies: EMBASE searches for surrogate endpoints and meta-analyses/network meta-analyses, supplemented by ChatGPT queries (with manual verification), Google Scholar searches, and review of myelofibrosis clinical guidelines and relevant NICE appraisals. Publications were screened for evidence on associations between post-treatment changes in intermediate clinical outcomes and overall survival.
RESULTS: A total of 11 publications (9 unique studies) were identified from EMBASE, with an additional 4 relevant studies found through ChatGPT-assisted searches. These included clinical trials and real-world studies highlighting associations between SVR (10% to 50% spleen volume reduction), BMF, transfusion status, anemia response, and OS. SVR was shown to be predictive of survival outcomes in several studies, both in JAK-inhibitor-naïve and -experienced populations. It was also identified as a key factor in multiple prognostic models and observational cohorts. Two network meta-analyses involving 9 randomized controlled trials incorporated studies reporting intermediate endpoints (SVR, anemia-related metrics) and OS outcomes, providing a potential evidence base for future surrogate endpoint validation analyses.
CONCLUSIONS: The review identifies potential surrogate endpoints - in particular SVR, and transfusion status - associated with predicting OS in patients with myelofibrosis. SVR emerges as a particularly relevant surrogate, given its strong association with disease burden and patient outcomes. These findings underscore the importance of these surrogate endpoints when considering the clinical benefit of treatment for patients with MF. Further validation using multivariate meta-analysis is recommended to quantify the strength of the surrogate relationship with OS.
Conference/Value in Health Info
2025-11, ISPOR Europe 2025, Glasgow, Scotland
Value in Health, Volume 28, Issue S2
Code
CO247
Topic
Clinical Outcomes, Study Approaches
Topic Subcategory
Relating Intermediate to Long-term Outcomes
Disease
Oncology