OBJECTIVES: Decision makers are frequently faced with the question of how realistic are cost-effectiveness models. Overall survival (OS) is a desired clinical trial and health technology assessment endpoint but frequently unrealistic or elusive at the point of new drugs financing decision. We aim to compare the overall survival from lenalidomide treatment in relapsed/refractory multiple myeloma (rrMM) patients estimated from modelling clinical trial data in a cost-effectiveness analysis (CEA) and compare it with real-word data (RWD) from South Korea. METHODS: In this CEA of lenalidomide-plus-dexamethasone OS outcomes were indirectly estimated using a quantitative relationship between time-to-progression/progression-free-survival derived with a censored normal weighted Tobit regression model with data from a literature review of 153 studies containing 230 treatment arms and 22,696 MM patients. Real world data was from a retrospective analysis of 110 heavily pre-treated patients from 20 hospitals.  These patients were treated with lenalidomide-plus-dexamethasone in the Korean patient access program between 2008 and 2012. RESULTS: The RWD was available for a heavily pre-treated population with 76.3% of patients with >2 previous treatments. For this population, estimated median time to progression (TTP) was 7 and 7.4 months in RWD and MM009/010 clinical trials, respectively. Corresponding predicted Tobit model upfront OS estimates, corrected for baseline characteristics, were 21.3 and 22.2 months. In the RWD data from South-Korea the reported median OS was 21 months. Median OS was also 21.0 months in the equivalent population of the two large, multicenter MM009/010 randomized phase III trials. CONCLUSIONS: This study provides evidence that overall survival of Lenalidomide-plus-dexamethasone treatment in relapsed/refractory multiple myeloma estimated in a cost-effectiveness analysis is remarkably similar to the outcomes observed in real-word patients and points to the value of such upfront estimates to decision makers in the absence of mature overall survival data.