BACKGROUND INFORMATION: Health Technology Assessment (HTA) is based on evidence resulting from randomized clinical trials and observational studies. In oncology, progression free survival and overall survival are usually the major endpoints for assessing efficacy. Treatments are almost always compared using survival model (e.g. the semi-parametric Cox model) adjusting for key prognostic factors. Inferences are therefore based upon adjusted treatment effects, expressed as adjusted hazard ratios and p-values. However, non-parametric Kaplan-Meier unadjusted survival curves are still routinely displayed along with their unadjusted survival medians. Hence, in some situations, adjusted statistics could contradict the apparent results observed with unadjusted survival curves. A literature review was conducted to evaluate the added value of adjusted statistics compared to unadjusted ones, regarding treatment baseline imbalances and confounding bias, especially in Real-World-Data studies coping with growing HTA expectations. DISCUSSION: Statistical methods are proposed in the literature [Nieto 1996] to adjust survival curves and provide adjusted difference in survival medians to match treatment effects estimated in a survival model accounting for potential confounders. A review of these methods along with their comparative merits and limitations will be presented. Applied literature on these methods seems to be rare since this kind of adjustment is only emerging. Thus, in application files, HTA do not systematically receive this adjusted information and make their decisions upon different sources of evidence which may be more or less inconsistent. Further discussion between methodologists, clinicians and Health Technology Assessors on this topic should be promoted so that trials and studies routinely provide key statistical results for making informed decisions. Finally, the authors’ recommendations to support HTA appraisal are to provide adjusted graphical displays of survival curves and adjusted differences in addition to unadjusted survival medians.