OBJECTIVES

: Conducting cost-effectiveness analyses of pan-tumor medications versus standard of care (SoC), based on basket, single arm trial data may present challenges due to the heterogeneity in outcomes across SoC within and between tumor types. Here we propose an approach to develop SoC estimates to conduct a cost-effectiveness analysis by using a weighted survival method (WSM) of published estimates.

METHODS

: Using a partition survival 3 state model, the Kaplan-Meier curves of the pan-tumor basket trial can be extrapolated using standard methods and area under the curve (AUC) was calculated to assess the survival in progression-free and progressed states. A targeted literature review was performed to collect the overall survival (OS) and progression-free survival (PFS) estimates for each tumor type and treatment line assessed in the basket trial, for the SoC. Reported median PFS and OS from literature was converted to mean values, based on exponential extrapolation assumption, to estimate the SoC AUC. The mean PFS and OS of each tumor type from the literature were then averaged and weighted by the proportion of tumor types in the basket trial cohort to simulate what the survival benefit of the SoC would have been in the basket trial population.

RESULTS

: OS and PFS median estimates from the literature for the SoC therapies ranged from 2.9 to 30.0 months. After applying WSM, the mean PFS and OS of the SoC cohort was estimated to be 7.9 and 22.5 months respectively.

CONCLUSIONS

: The WSM approach to compare AUC mean survival may be a feasible approach to assessing the cost-effectiveness of a pan-tumor medication, however this approach is limited by the size of the basket trial cohort, heterogeneity in the SoC per tumor and inability to adjust for patient characteristics.