OBJECTIVES: Typically, health technology assessment (HTA) focuses on overall survival (OS) as the key clinical endpoint when making decisions on new treatments in oncology. However, capturing OS in early-stage breast cancer (eBC) is not always possible, as longer follow-ups are required. Therefore, this study assessed patient preferences for treatment attributes and endpoints in neoadjuvant therapy where mature and non-confounded OS is unlikely to be available.

METHODS: An online survey was developed with healthcare professionals and patient advisory groups and completed by HER2+ eBC patients in Germany, France, Italy and Spain. Patients were presented with 15 discrete choice experiment (DCE) tasks and for each task, asked to choose between two hypothetical treatment profiles, plus an opt-out option. Each DCE profile included 5 attributes with accompanied definitions: OS at 5-years (%) (5-levels), disease-free survival (DFS) at 5-years (%) (5-levels), pathological complete response (pCR) (%) (4-levels), impact of side effects on quality of life (QoL) (3-levels), and ability to receive breast conserving surgery (2-levels). Data was analysed using multinomial logit regressions.

RESULTS: 334 HER2+ eBC patients completed the survey. Patients placed most importance on increases in pCR from 25% to 100% (coefficient = 0.55, odds ratio = 1.73, p-value < 0.001), followed by increases in DFS from ‘data unknown’ at 5-years to 95% (coefficient = 0.39, odds ratio = 1.48, p-value < 0.001). Based on the preference weights, pCR (31%) was most important relative to all other attributes, followed by DFS (24%), and OS (22%).

CONCLUSIONS: pCR and DFS were considered important by HER2+ eBC patients. These findings are in-line with EMA and FDA reviews that accept pCR as a clinically valid endpoint. It is important HTA bodies and payers consider patient preferences and the relative importance placed on different treatment attributes and endpoints in their decision-making.