Presentation (CTI)

OBJECTIVES: Chimeric antigen receptor T-cell therapies (CAR-Ts) have emerged as promising treatments for relapsed or refractory hematologic malignancies. They are usually associated with a high up-front cost and a complex manufacturing and administration process. This study aims to explore the approaches taken to model the cost effectiveness (CE) of these therapies in drug submissions to UK’s National Institute for Health and Care Excellence (NICE) and Canada’s Drug Agency (CDA-AMC).
METHODS: All publicly available NICE and CDA-AMC model appraisal reports up to June 2024 were identified from which information on CE model structures, inputs, underlying assumptions, model outcomes and evidence review group critiques were extracted and assessed.
RESULTS: Seventeen appraisals were identified (NICE: 7, CDA: 10). All models were 3-health-states Partitioned Survival Models (PSMs), six with a precedent decision-tree, and twelve included a mixture-cure component to extrapolate the overall survival (OS) to reflect potential cure for a patient subset. The submitted prices ranged from £282,000 to £297,000, and from C$485,021 to C$632,455 per administration, resulting in incremental cost-effectiveness ratios (ICERs) in the range of £20,000 to £60,000, and from C$58,178 to C$216,606 per quality-adjusted life year (QALY), for UK and Canada respectively. For England a single tariff to capture the costs of delivering CAR-Ts of £58,964 has now been established. Uncertainty in the OS long term survival benefit was the key limitation discussed across all appraisals. Despite the methodological challenges, most appraisals (15 of 17) led to positive reimbursement decisions, conditional on price reductions, collection of additional data and/or restricted to administration at specialized centres.
CONCLUSIONS: Standard economic modelling approaches in oncology (PSMs) were deemed adequate by NICE and CDA-AMC to inform reimbursement decisions for CAR-Ts. Further research into the long-term benefits of CAR-Ts is needed to determine whether new, non-conventional modelling approaches are warranted to accurately capture their value.