Updated cost-effectiveness analysis of trastuzumab for early breast cancer:A UK perspective considering long-term toxicity and pattern of recurrence.

Background: Trastuzumab has significantly improved survival outcomes for women with HER2 positive early breast cancer. Trastuzumab was established as a costeffective adjuvant treatment in 2006. We present an updated cost-effectiveness analysis from the UK perspective which explores assumptions about the duration of benefit from treatment, pattern of metastatic recurrence and long-term cardiac toxicity. Methods: A cost-utility analysis was performed using a discrete-state timedependent semi-Markov model to calculate expected costs and benefits over the lifetime of an average cohort of women with HER2 positive early breast cancer treated with or without one year of adjuvant trastuzumab sequentially after chemotherapy. The perspective was the UK NHS. Probabilistic sensitivity analysis is used to characterise uncertainly around expected outcomes. Value of information analysis is used to identify areas of priority for further research. Results: The cost-effectiveness estimates are highly sensitive to the estimated duration of treatment benefit. Trastuzumab remains the cost-effective treatment strategy at a willingness-to-pay threshold of £30,000 per QALY provided the duration of benefit is more than 3.6 years from treatment initiation. An increasing proportion of brain metastases with trastuzumab produces small reductions in the costeffectiveness estimate. Long-term cardiac toxicity must rise to high levels to affect overall life-expectancy and cost-effectiveness. VoI analysis places highest value on research into the duration of treatment benefit. The relationship between progression free survival and overall survival and the costs of cancer recurrence are also important. Conclusion: The use of adjuvant trastuzumab remains cost-effective given current estimates of the duration of treatment effect. Uncertainty around cost-effectiveness could be reduced further by research into the duration of treatment effect, particularly in subgroups where this may be shorter. Long-term follow-up is warranted and methods to accurately measure duration of treatment effect and late toxicities should be developed for future adjuvant drug studies.