Objective: The introduction provisional approval strategies increases the approval of anticancer drugs with ambiguous benefit-risk profiles. Thus, in many instances, there is lacking evidence about overall survival (OS) at the time of marketing authorisation. Our objective was to monitor and characterise therapies with ambiguous benefit-risk profiles and identify any postapproval updates on median OS after at least 3 years of approval by the European Medicines Agency (EMA).
Methods: We included all originator anticancer drugs with initially ambiguous benefit-risk profiles that received marketing authorisation by the EMA between January 1, 2009 and May 31, 2015. Our monitoring timeframe was at least 3 years after EMA approval. To identify study updates, the following three sources were included: clinicaltrials.gov, European Public Assessments Reports and PubMed.
Results: In total, we identified 102 eligible approval studies. Out of these, a negative difference in median OS or no information was available in 43 (42.2%) instances. During monitoring, 14 updates with accessible positive information on OS could be identified. Including monitoring results, there are still 29 remaining therapies (28.4%) where no or negative information (n = 24 [23.5%] and n = 5 [4.9%], respectively) regarding OS is present at least 3 years after EMA approval.
Conclusion: One-third of oncology drugs with ambiguous benefit-risk profiles at the time of approval fail to demonstrate a survival benefit even after several years of marketing authorisation. Systematic and transparent postapproval monitoring mechanisms will be of high relevance to assure a clinically relevant patient benefit, since the trend towards faster access to medicines with uncertain benefit is increasing rather than declining.
Keywords: Additional monitoring; Benefit-risk profile; Cancer treatment; Conditional marketing authorisation; Drug approval; EMA; Orphan drug; Provisional approval; Survival.
Copyright © 2019 The Author(s). Published by Elsevier Ltd.. All rights reserved.