OBJECTIVES: Due to improvements in survival time with new cancer therapies and the need for lengthy follow-up, median overall survival (OS) data are scarce. The use of alternative clinical endpoints may facilitate earlier analysis of trial data. However, there are challenges with the acceptance of these endpoints by health technology assessment (HTA) agencies. Our aim was to evaluate alternative endpoints to median OS in advanced melanoma, using methodology acceptable to HTA agencies, including IQWiG.

METHODS: A systematic literature review (SLR) was conducted to identify randomised clinical trials (RCTs) of immunotherapies or targeted therapies in advanced melanoma, reporting median OS. A multi-trial statistical approach was used to calculate arm-level (R_A) and trial-level (R_T) correlation coefficients between alternative endpoints and median OS. Significant results led to the calculation of the surrogate threshold effect (STE) (minimum treatment effect on the alternative endpoint required to expect a benefit in median OS), and a number of sensitivity analyses were conducted.

RESULTS: The SLR identified 25 RCTs for inclusion in the base case analysis and 25 candidate alternative endpoints. Base case analysis found OS12m (12 month survival rate) (R_A