Visualization of Multi-indication Randomized Control Trial Evidence to Support Decision Making in Oncology: A Case Study on Bevacizumab

BackgroundAs an increasing number of oncology drugs are licensed for multiple indications, sharing information across indications may help improve the precision of estimates for a target indication where evidence may be immature. Visualizing the accumulation of evidence and its characteristics across all indications can help inform policy makers as to whether multi-indication synthesis methods should be considered and guide expert elicitation on appropriate cross-indication assumptions.MethodsThe multi-indication oncology drug bevacizumab was selected as a case study. We used visualization methods including timeline, ridgeline, and split-violin plots to display evidence and synthesis results across 7 licensed cancer types, focusing on the evidence on overall and progression-free survival and the display of results from models with and without information sharing.ResultsThe proposed displays allow for visualization of key characteristics of the evidence to support the assessment of heterogeneity within and across indications and inform the feasibility of information-sharing models.LimitationsThe lack of consistent reporting of data in trial reports limits the visualization of some study characteristics. Tradeoffs between plot readability and the level of detail to include were required.ConclusionsClear graphical representations of the evolution and accumulation of evidence and synthesis results can provide a better understanding of the entire multi-indication evidence base, which can inform judgments regarding the appropriate use of data within and across indications. Interactive plots could help overcome some of the current limitations.ImplicationsThe proposed displays should be used to facilitate discussion with experts on the judgments required to assess the feasibility of using information-sharing methods to improve the estimation of relative treatment effects in evidence synthesis approaches and health technology assessment.HighlightsAn increasing number of oncology drugs are licensed for multiple indications; we developed visualization methods for multi-indication evidence that consider key characteristics unique to oncology.Graphical displays can be used to show the evolution of evidence within and across multiple indications.Clear evidence visualizations can be used as a tool to support evidence synthesis approaches, support policy makers, or guide expert elicitation.