Evolving Assumptions on Treatment Durability in Cost-Effectiveness Models of Cell and Gene Therapies: Insights From French HTA Submissions
Author(s)
Rares Enache, pharmD, Romain Moreau, PharmD, Stève Bénard, PharmD, Elise Cabout, MSc, Nassim Ahmed, PharmD.
stève consultants, Oullins, France.
stève consultants, Oullins, France.
OBJECTIVES: Cell and gene therapies (CGTs) hold promise for long-term disease modification, especially in high-burden conditions. However, limited long-term data challenges cost-effectiveness evaluations. As treatment durability (i.e. duration of treatment effect [TE]) is a key value driver, this study examines how the French HTA commission for health economics assessment (CEESP) has approached long-term modeling assumptions over time.
METHODS: We reviewed treatment durability assumptions, modeling methods and CEESP’s opinions for 11 CGTs evaluated between 2019 and 2025. For products reassessed for the same indication, only the most recent evaluation was included.
RESULTS: Since 2019, the CEESP assessed 15 models across various therapeutic areas. 80% raised methodological concerns about TE extrapolation and only half were validated. The median time horizon used in base-case was 20 years (from 8 to 85 years), while the median pivotal trial follow-up was only 1.5 years. Yet only one-third of submitted models included TE waning. From 2019-2022 to 2023-2025, the median time horizon in CEESP-validated models increased by 58%, showing greater openness to long-term projections. However, this was associated with a 28% reduction in the duration of full TE, indicating a shift toward more conservative treatment durability assumptions. This trend is reinforced in the six most recent models (Nov 2023-Feb 2025). The only two validated models both included TE waning starting at one-third of the time horizon suggesting that CEESP favors models that combine long-term projections with realistic treatment durability assumptions. Omitting waning may compromise model credibility—especially when the ICER is dominant.
CONCLUSIONS: While the CEESP has become more accepting of extended time horizons in CGT models, this has been counterbalanced by a growing expectation for TE waning in main analyses. Recent assessments suggest that incorporating waning assumption is critical for model validation, reflecting a more cautious stance on long-term efficacy projections.
METHODS: We reviewed treatment durability assumptions, modeling methods and CEESP’s opinions for 11 CGTs evaluated between 2019 and 2025. For products reassessed for the same indication, only the most recent evaluation was included.
RESULTS: Since 2019, the CEESP assessed 15 models across various therapeutic areas. 80% raised methodological concerns about TE extrapolation and only half were validated. The median time horizon used in base-case was 20 years (from 8 to 85 years), while the median pivotal trial follow-up was only 1.5 years. Yet only one-third of submitted models included TE waning. From 2019-2022 to 2023-2025, the median time horizon in CEESP-validated models increased by 58%, showing greater openness to long-term projections. However, this was associated with a 28% reduction in the duration of full TE, indicating a shift toward more conservative treatment durability assumptions. This trend is reinforced in the six most recent models (Nov 2023-Feb 2025). The only two validated models both included TE waning starting at one-third of the time horizon suggesting that CEESP favors models that combine long-term projections with realistic treatment durability assumptions. Omitting waning may compromise model credibility—especially when the ICER is dominant.
CONCLUSIONS: While the CEESP has become more accepting of extended time horizons in CGT models, this has been counterbalanced by a growing expectation for TE waning in main analyses. Recent assessments suggest that incorporating waning assumption is critical for model validation, reflecting a more cautious stance on long-term efficacy projections.
Conference/Value in Health Info
2025-11, ISPOR Europe 2025, Glasgow, Scotland
Value in Health, Volume 28, Issue S2
Code
HTA141
Topic
Economic Evaluation, Health Policy & Regulatory, Health Technology Assessment
Topic Subcategory
Value Frameworks & Dossier Format
Disease
Genetic, Regenerative & Curative Therapies