METHODS: Women in the United States with self-reported physician diagnoses of ovarian cancer and eligible for maintenance therapy completed an online discrete-choice experiment (DCE) survey. The survey presented nine choice questions, each including a pair of hypothetical medicine profiles with varying efficacy, tolerability, and risks of side effects. Each profile was defined by the following attributes identified from the literature with clinical input and tested in patient interviews: progression-free survival (PFS), fatigue, diarrhea, nausea and vomiting, hypertension, and risk of gastrointestinal (GI) perforation. The profiles in the choice questions were based on an experimental design with known statistical properties. Random-parameters logit was used to estimate preferences.

RESULTS: Two hundred women completed the survey; median age  was 49 years, 26% were late stage (3/4), and 44% had been diagnosed within 2 years. Across the attributes, better outcomes were significantly preferred to worse outcomes, except that respondents did not distinguish between no nausea and mild nausea. Relative to the other attributes and levels, respondents placed the greatest weight on avoiding severe diarrhea, followed by reducing the risk of GI perforation, and increased PFS. Respondents were willing to give up 6.5 months of PFS to reduce diarrhea from severe to none. No statistical differences were found between the overall preferences of early versus late stage respondents, respondents above and below the median age in the sample, and respondents who had been diagnosed in the last 2 years compared to those diagnosed more than 2 years ago.

CONCLUSIONS: Women with ovarian cancer were willing to trade-off efficacy (PFS) for improvements in side effects and risk. The lack of differences across subgroups suggest consistent preferences across the attributes within our sample. Funded by GSK.