OBJECTIVES: Decision makers are increasingly requiring both cost-effectiveness (CE) and affordability information of new technologies, which has resulted in the use of CE thresholds (CET) and budget impact thresholds (BIT) in many countries. Cost-effectiveness affordability curves (CEAfCs), a tool to present the joint probability of CE and affordability of a healthcare intervention at varying CETs and BITs, have been used in economic evaluations of preventive interventions in developing countries. The objective of this study is to apply CEAfCs to the evaluation of cancer drugs in England and understand the added value of this methodology in this setting.

METHODS: A combined cost-effectiveness and budget impact analysis model from a payer perspective was built. A three-state Markov was populated using published data on trastuzumab emtansine (T-DM1) vs lapatinib-capecitabine (LC) for HER2+ metastatic breast cancer (mBC) in England. Probabilistic sensitivity analyses were performed. By combining the commonly used cost-effectiveness acceptability curves (CEACs) with the more rarely-used affordability curves (ACs), CEAfCs were produced. The CEAfCs at different price levels were explored.

RESULTS: Due to the small number of treated patients, the budget impact of T-DM1 was below the £20 million threshold. However, the model found that T-DM1 was not cost-effective at its list price compared with LC. Assuming a value-based price using the CET of £50,000 per QALY for end-of-life treatment, the joint probability of T-DM1 being both cost-effective and affordable greatly increased. CEAfCs were produced based on different assumptions affecting CE and/or affordability and presented in a single diagram.

CONCLUSIONS: As increasingly stricter CET and BIT are applied globally, CEAfCs can be a useful tool to inform decision-makers about the probability of an intervention being both cost-effective and affordable. Communication of uncertainty in CEAs and BIAs would benefit from increased use of CEAfCs as a complementary analysis to CEACs and ACs.