OBJECTIVES: Molecularly targeted therapies such as cetuximab and bevacizumab have improved survival for patients with metastatic colorectal cancer (mCRC) significantly. Understanding patient preferences for these treatments as well as their willingness-to-pay is crucial in shared decision making. Discrete Choice Experiment (DCE) is increasingly used in healthcare to elicit preferences as it can simulate real-world trade-offs between attributes. This study aims to design a DCE to measure patient preferences and willingness-to-pay for targeted therapies for mCRC. METHODS: In stage 1, literature review and patient interviews were conducted to identify and define important attributes and levels. In stage 2, a draft DCE questionnaire designed based on patients’ feedback was pilot-tested. Patients with stages 2 to 3 CRC were asked to choose one out of two hypothetical drug alternatives, in each of the five choice tasks in the questionnaire. Feedback on pictorial representation, labelling and framing of attributes and dual-response none option were solicited.  RESULTS: In Stage 1 (n=6), all except one chose progression-free-survival as being more important than overall survival. All except two preferred full description of severity of skin rashes over mere labelling as “mild”, “moderate” and “severe”. In Stage 2, 13 patients completed the draft questionnaire, comprising six attributes (progression-free survival, severity of acne-like skin rashes, severity of bleeding, out-of-pocket cost, frequency of administration and degree of independence in medical decision making) with varying attribute levels (2 to 4 levels). Patients prefer out-of-pocket cost to be presented as total cost over treatment lifetime rather than cost-per-month without a defined duration. Patients seemed to have difficulty answering the dual-response none (“Will you really buy the medication you have chosen? Yes or No?”) option.  CONCLUSIONS: Initial lack of clarity in the DCE questionnaire was overcome. Patients were able to complete the choice tasks with little help. A full-scale study is currently in progress.