OBJECTIVES: Health state utility data used in economic models is critical in health technology assessment (HTA) of new medical innovations. Utility data is typically collected alongside clinical trials. In oncology trials patients are often assessed regularly prior to disease progression, whereas post-progression follow-up is usually limited. Data gaps in trial-based utility data lead to mixing trial- and literature-based utilities in the economic models, which could introduce inconsistencies and extra uncertainty into the model. In this study, we focused on multiple myeloma (MM) and investigated the utility assumptions across HTA submissions.

METHODS: A targeted literature review was conducted to identify the HTA submissions in patients with MM with at least one or two prior lines of therapy (1PL+, 2PL+). The search included NICE, ICER, CADTH, SMC and PBAC submissions. All utility values along with their sources were extracted and reported.

RESULTS: Twelve submissions reported utility values across different HTAs. The variation in utility drop upon progression was large: In patients with exactly 1PL, the utility drop upon progression was ranging between 4% and 26% of pre-progression. This range was 6%-17% for patients with exactly 2PL, 4%-21% for patients with 1PL+, and 6%-34% for patients with 2PL+. In most submissions, pre-progression utilities were collected from clinical studies while post-progression utilities were sourced from literature or real-world data (RWD). Literature and RWD sources were criticised by HTAs due to differences in patients’ populations compared to trials.

CONCLUSIONS: We observed large variations in the utility decline upon progression across different MM HTA submissions. This implies that alternative assumptions could have been used, with a potential large impact on the ICERs. Further work is required to identify accurate utility values for post-progression, e.g. by identifying sources with similar population to the post-progression population, or by introducing adjustment factors to account for the differences in populations.