Frank Xiaoqing Liu, Edward A Witt, Scot Ebbinghaus, Grace DiBonaventura Beyer, Reshma Shinde, Enrique Basurto, Richard W Joseph

To examine and compare patient and oncologist preferences for advanced melanoma treatment attributes and to document their trade-offs for benefits with risks. A discrete choice experiment (DCE) was conducted among advanced melanoma patients and oncologists. Qualitative pilot testing was used to inform the DCE design. A series of scenarios asked stakeholders to choose between two hypothetical medications, each with seven attributes: mode of administration (MoA), dosing schedule (DS), median duration of therapy (MDT), objective response rate (ORR), progression-free survival (PFS), overall survival (OS), and grade 3-4 adverse events (AEs). Hierarchical Bayesian logistic regression models were used to determine patients' and oncologists' choice-based preferences, analysis of variance models were used to estimate the relative importance of attributes, and independent t-tests were used to compare relative importance estimates between stakeholders. In total, 200 patients and 226 oncologists completed the study. OS was most important to patients (33%), followed by AEs (29%) and ORR (25%). For oncologists, AEs were most important (49%), followed by OS (34%) and ORR (12%). An improvement from 55% to 75% in 1-year OS was valued similar in magnitude to a 23% decrease (from 55% to 32%) in likelihood of AEs for oncologists. Patients valued OS, AEs, and ORR sequentially as the most important attributes in making a treatment decision, whereas oncologists valued AEs most, followed by OS and ORR. In comparison, patients differed significantly from oncologists on the importance of ORR, AEs, and PFS, but were consistent in OS and the rest of attributes.