Background: Most patients with metastatic breast cancer are treated with chemotherapies and/or targeted therapies. These therapies have toxicity profiles that vary with agent(s) used. Patient attitudes towards different adverse events (AE's) may vary and factor into treatment decisions. Different patients may have specific feelings about tolerable and unacceptable AE's, especially when balanced against possible treatment benefit. As more agents/combinations become available, it becomes increasingly important to understand which adverse events impact treatment decisions. Previous research has shown that conjoint analysis (CA) is a valid methodology that allows patients to express preferences and is particularly useful when designed based on specific treatment profiles (Smith, ASCO 2011; Smith, ASCO 2012).

Methods: The objective of this study was to assess patient preferences using CA based on profiles of two MBC regimens (trastuzumab+docetaxel and T-DM1). Patients were presented pairs of hypothetical treatments (describing benefit, AE's, and cost) and asked what their preferred alternative was; a follow-up question asked if they would take the treatment if it were the only option available. Five AE's (alopecia, peripheral neuropathy, diarrhea, fatigue, and neutropenia) with differing likelihood, severity, and/or duration were included. There were 3 stages in preparing the CA survey: the first comprised two online focus groups conducted with patients with metastatic disease. Stage 2 included the development of the CA survey using patient language to describe the AE's and their impact, as well as images to represent likelihoods, progression-free survival (PFS), and costs. Stage 3 is initiating and will recruit patients with the assistance of several breast cancer organizations (target n= 600). Analysis of response patterns allows study of the influence of each variable and provides a basis for prediction of treatment choice for any combination of benefit, AE's, and cost. Final analysis will be complete in September 2012.

Results: Findings from the focus groups facilitated an understanding of PFS, of experience with and impact of the AE's on decision-making, and of attitudes. In Stage 2, the survey was pretested with seven patients and took approximately 20 minutes to complete. Feedback suggested the questions were relevant and realistic. Suggestions allowed for improvement of the CA explanatory material, as well as refinement of a few answer choices to questions outside the CA section of the survey. Final study results will present the proportion of patients who are predicted to prefer each of two treatment profiles, the impact of each attribute level on treatment preference and differing preferences seen in patient subgroups. This information will provide valuable insight into patient preferences and inform future development of new therapies. In addition, these results may generate discussion and consideration of patient preferences in conversations about patient care and treatment selection.

Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P6-09-11.