항암제의 경제성평가 모델링을 위한 혼합 치료 모형의 적용 현황 및 향후 적용 방향

Abstract

Objectives: The aim of this study was to review the theoretical foundations and practical applica- tions of the mixture cure model (MCM) in Health Technology Assessment (HTA), particularly in the economic evaluation of oncology drugs. We focused on the context of the UK, where MCM has gained increasing attention within National Institute for Health and Care Excellence (NICE) ap- praisals. Methods: We conducted a narrative literature review of published methodologies on MCM, including technical support documents from the NICE Decision Support Unit. We then sys- tematically reviewed NICE technology appraisal guidance published between January 2020 and February 2025, identifying cases in which MCM was applied as part of survival extrapolation for oncology drugs. For each case, we extracted the rationale for using MCM, the type of therapy and cancer, and the committee’s assessment of model appropriateness. Results: Out of 414 appraisals, 10 used MCM as part of the base-case analysis for long-term survival extrapolation. These primarily included chimeric antigen receptor T-cell therapies, immune checkpoint inhibitors, and targeted agents in both hematologic and solid tumors. The main justifications for MCM use were observed plateaus in survival curves, long-term survival evidence, and clinical expert consensus. However, concerns remained regarding the face validity of MCM when based on limited follow-up or sparse data. While earlier cases (e.g., in 2020) saw MCM rejected due to uncertainty, more recent apprais- als (2023–2024) show growing institutional acceptance of MCM, especially when supported by real- world evidence or mature clinical data. Conclusion: The application of MCM in oncology drug evaluations reflects a methodological evolution in HTA. As the number of therapies offering cura- tive potential increases, MCM provides a flexible and realistic framework for survival modeling. However, its application must be accompanied by sensitivity analyses and cautious interpretation, especially in the presence of limited follow-up. These findings support the need for future method- ological guidance and policy discussions on the use of MCM in HTA.