OBJECTIVES: Cost-effectiveness studies in early cancer typically require more complex model structures than for advanced disease due to the need to capture the downstream consequences of treatment in avoiding or delaying the burden of advanced cancer. The development of these models can be time-consuming and may be prone to error, requiring specialized programming. Our aim was to develop an open-source framework that can be used to model outcomes across multiple early cancer settings.

METHODS: A review of decision models used in health technology assessment submissions of treatments in early-stage cancer was performed. Key modelling features were identified and summarized across indications. An Excel-based template model was created incorporating common features of existing models.

RESULTS: We present an open-access model template, based on a Semi-Markov state transition model containing four core health states of event-free, non-metastatic/locoregional recurrence, distant metastatic (DM) and death. Within the DM state, the proportions of patients who are progression-free and progressed are modelled using a nested partitioned survival approach. Additional sub-states for ‘early’ and ‘late’ distant metastases provides flexibility to model different pathways and outcomes based on time of recurrence. All transition risks are dependent on time since entry to state and modelled via tunnel states. Input parameters for the template model include the survival parameters fitted to each cause-specific hazard, age- and gender-specific mortality lifetables, and a standardized mortality ratio for modelling excess mortality. Results include projected state-occupancy over time, and mean times in state, and life years. The template contains guidance on how to apply the model to future cost-effectiveness studies.

CONCLUSIONS: The model template is designed to support future cost-effectiveness studies in early cancer, reducing the burden of de-novo model build and potentially improving consistency and transparency in the health state modelling of early cancer treatment for health technology developers and decision-makers.