OBJECTIVES : Immune checkpoint inhibitors (ICIs) in oncology are often associated with large uncertainty surrounding long-term overall survival (OS) and progression-free survival (PFS), which are often key drivers of cost-effectiveness. A conservative approach to address the uncertainty of the long-term treatment effect on OS and PFS is to assume treatment effect waning. This is opposed to assuming that the treatment effect observed in the trial data continues over a lifetime. This study aims to identify and review these methods in technology appraisals (TAs) submitted to the National Institute for Health and Care Excellence (NICE).

METHODS : The inclusion criteria for treatment effect waning was defined as any instance where the therapeutic benefit of the intervention diminished or was capped in the analysis. TAs for ICIs published between 2014 and 2019 were identified where treatment effect waning was modelled in either the company submission or the Evidence Review Group (ERG) revised model.

RESULTS : Seventeen TAs were identified. Five TAs presented waning in the company’s base case while nine presented waning in scenarios only. Waning was not presented in the manufacturer’s submission in three TAs, but it was explored by the ERG. The method of setting the hazard ratio to 1 for the intervention relative to the comparator for OS and/or PFS was consistent across all identified submissions. This was generally implemented after 3-5 years. Waning was applied to OS in most cases, however, three submissions applied waning to PFS only. ERG criticisms of treatment effect assumptions included clinical implausibility of a lifetime treatment effect and the timepoint at which a waning effect cap was applied.

CONCLUSIONS : The method and rationale for modelling treatment effect waning across ICI oncology TAs is generally consistent to address the uncertainty and potential overestimation of long-term treatment benefits in reimbursement decision making.