Sensitivity Analysis in Economic Evaluations of Immuno-Oncology Drugs: A Systematic Literature Review

Abstract

ObjectivesThis study aimed to review, assess, and report the characteristics and strategies of sensitivity analyses (SAs) that were performed in the context of published economic evaluations of immuno-oncology drugs. MethodsThe systematic literature search was conducted in Scopus and MEDLINE for articles published from 2005 to 2021. Study selection, based on a predefined set of criteria, was performed by 2 reviewers independently. We included economic evaluations of Food and Drug Administration–approved immuno-oncology drugs that were published in English and assessed the accompanying SAs on a set of items, including the range justification of the baseline parameters within the deterministic SA, the provisions for the correlation/overlay between parameters, and the justification of the chosen parameter distribution for the probabilistic SA, among others. ResultsA total of 98 of 295 publications met the inclusion criteria. A total of 90 studies included a one-way and probabilistic SA and 16 of 98 studies had one-way and scenario analysis, alone or together with probabilistic analysis. Most studies provide explicit references as to the choice of parameters and values; nevertheless, there is a lack of a reference of correlation/overlay between parameters in most of the evaluations. In 26 of 98 studies, the most influential parameter for the incremental cost-effectiveness ratio was the under-evaluation drug cost. ConclusionsMost of included articles contained an SA that was implemented according to commonly accepted published guidance. The under-evaluation drug cost, the estimates of progression-free survival, the hazard ratio for overall survival, and the time horizon of the analysis seem to play an important part in the robustness of the outcomes.