OBJECTIVES: In cancer, treatments often aim to extend time to progression. The implications on overall survival are often inconclusive, as trials are too short and the majority of patients are still alive at the end of the trial. However, for decision making, it is important to estimate both the treatment effect on Progression Free Survival as well as the treatment effect on Overall Survival. This poster shows how the estimation of Overall Survival benefit can be improved by the use of Progression Free Survival data. METHODS: The developed Network Meta-Analysis model uses the tested hypothesis that treatments provided until progression in general do not change the length of the post-progression period. This hypothesis is tested in detail based on systematic literature reviews concerning 4 different types of cancer. A test for equal lengths of the post-progression periods is described too. RESULTS: A network meta-analysis model is described, which can be used to obtain estimates for OS from PFS data for treatments for which no OS data or insufficient OS data are available. This informs decision making in situations where otherwise no conclusion can be drawn. The methodology is applied to an indirect comparison of Chlorambucil, Fludarabine and Fludarabine Cyclophosphamide. Comparable DIC were obtained to individual fitting of OS and PFS for the situation that OS data were available. Therefore, the methodology both enables the fitting of OS when OS data are not available as well as potentially improves OS fitting when data are limited available. CONCLUSIONS: Based on systematic literature reviews, a method is developed to use PFS as surrogate outcome for OS. In addition, a test is developed to justify the assumption of equal post-progression periods among treatments, which can be used to assess whether the translation of PFS time differences in OS time differences is appropriate.