OBJECTIVES:  Gene expression profiling (GEP) of tumours informs baseline risk prediction, potentially affecting decisions about adjuvant chemotherapy for women with early breast cancer (BrCa), of whom only 15% will experience a recurrence. Limited evidence exists on the clinical utility of GEP in chemotherapy treatment decisions. We aimed to measure the value of GEP testing information in chemotherapy treatment decisions based on risk-benefit tradeoffs from a stated preferences study. METHODS: Based on literature review, focus groups and interviews with BrCa patients and medical oncologists, we developed a discrete choice experiment survey. For our pilot, we surveyed BrCa patients (n=27), women from the general public (n=55), and medical oncologists (MOs; n=3) across Canada. The DCE included 12 choice tasks with 5 attributes and 3 scenario profiles considering orthogonality, D-efficiency and level balance. Preferences were analyzed using conditional logit and hierarchical Bayes and evaluated for goodness-of-fit. RESULTS: Most (>80%) respondents know someone who had chemotherapy for cancer. However, few respondents (<10%) know someone who had GEP testing. Across the three groups, the most important attributes in chemotherapy treatment decisions were (in order): GEP test score indicating likely benefit from chemotherapy, doctor’s estimate of risk of cancer returning (based on clinical algorithms), likelihood of permanent side effects, trust in cancer treatment doctor, and likelihood of temporary side effects.  In a scenario of intermediate risk of cancer returning based on clinical algorithms alone (no GEP score), 12% of respondents chose chemotherapy compared to 89% of respondents with a GEP score of 44 (high likelihood of benefit from chemotherapy). CONCLUSIONS: GEP testing is highly valued and strongly influences chemotherapy treatment decisions in all three groups. These findings provide preliminary evidence supporting the clinical utility of GEP in BrCa treatment decisions.