USING EARLY ECONOMIC EVALUATION TO IDENTIFY VALUE DRIVERS FOR ALGORITHM-GUIDED TREATMENT ALLOCATION IN ADVANCED HEPATOCELLULAR CARCINOMA (HCC): A HYBRID DECISION TREE-PARTITIONED SURVIVAL FRAMEWORK
Author(s)
Kang Wang, BSc, MSc1, Yirui Qian, BSc, MSc, PhD2, Huajie Jin, BSc, MSc, PhD1.
1Health Service & Population Research, Institute of Psychiatry, Psychology & Neuroscience (IoPPN), King's College London, London, United Kingdom, 2University of York, York, United Kingdom.
1Health Service & Population Research, Institute of Psychiatry, Psychology & Neuroscience (IoPPN), King's College London, London, United Kingdom, 2University of York, York, United Kingdom.
OBJECTIVES: Precision medicine approaches may improve treatment allocation in advanced hepatocellular carcinoma (HCC), where response to systemic therapies remains heterogeneous. However, the value of algorithm-guided treatment depends not only on predictive accuracy, but also on downstream treatment pathways, treatment duration, drug prices and survival gains. This study aimed to develop and apply an early economic modelling framework to identify key value drivers and evidence gaps for a precision medicine algorithm (PMA) designed to support systemic therapy allocation.
METHODS: A hybrid decision tree and partitioned survival model (PSM) was developed to compare PMA-guided allocation with Treatment as Usual. The decision tree linked diagnostic performance to treatment pathways using a positive prediction rate re-parameterisation approach Clinical inputs were derived from a retrospective multicentre cohort study (2015-2023; N=60). Comparator survival data were reconstructed from published trials. Costs, utilities, and adverse event parameters were obtained from published literature and UK reference sources. Outcomes included costs, quality-adjusted life-years (QALYs), and incremental cost-effectiveness ratios (ICERs).
RESULTS: In the base-case analysis, PMA increased costs by £90,733 and generated an additional 0.65 QALYs, resulting in an ICER of £140,401/QALY. When time-on-treatment was modelled separately, the ICER decreased to £27,613/QALY, highlighting substantial sensitivity to structural assumptions around treatment duration. Scenario analyses showed that outcomes were also strongly influenced by treatment allocation pathways and drug prices. Improvements in predictive performance did not necessarily translate into improved cost-effectiveness where they increased allocation to high-cost treatments without proportionate survival gains.
CONCLUSIONS: This early economic evaluation should be interpreted as a decision-support and evidence-generation exercise rather than a definitive assessment of cost-effectiveness. The framework demonstrates how early modelling can identify uncertainties, test value drivers and inform future study design, pricing considerations and evidence generation for precision oncology.
METHODS: A hybrid decision tree and partitioned survival model (PSM) was developed to compare PMA-guided allocation with Treatment as Usual. The decision tree linked diagnostic performance to treatment pathways using a positive prediction rate re-parameterisation approach Clinical inputs were derived from a retrospective multicentre cohort study (2015-2023; N=60). Comparator survival data were reconstructed from published trials. Costs, utilities, and adverse event parameters were obtained from published literature and UK reference sources. Outcomes included costs, quality-adjusted life-years (QALYs), and incremental cost-effectiveness ratios (ICERs).
RESULTS: In the base-case analysis, PMA increased costs by £90,733 and generated an additional 0.65 QALYs, resulting in an ICER of £140,401/QALY. When time-on-treatment was modelled separately, the ICER decreased to £27,613/QALY, highlighting substantial sensitivity to structural assumptions around treatment duration. Scenario analyses showed that outcomes were also strongly influenced by treatment allocation pathways and drug prices. Improvements in predictive performance did not necessarily translate into improved cost-effectiveness where they increased allocation to high-cost treatments without proportionate survival gains.
CONCLUSIONS: This early economic evaluation should be interpreted as a decision-support and evidence-generation exercise rather than a definitive assessment of cost-effectiveness. The framework demonstrates how early modelling can identify uncertainties, test value drivers and inform future study design, pricing considerations and evidence generation for precision oncology.
Conference/Value in Health Info
2026-11, ISPOR Europe 2026, Vienna, Austria
Value in Health, Volume 29, Issue 12S
Code
MSR145
Topic
Health Technology Assessment, Medical Technologies, Methodological & Statistical Research
Disease
Gastrointestinal Disorders, No Additional Disease & Conditions/Specialized Treatment Areas, Oncology, Personalized & Precision Medicine