Graphical representations of patient tolerability data: Recommendations from the National Cancer Institute (NCI) Cancer Moonshot Standardization Working Group.

Abstract

e18612 Background: Effective communication of treatment tolerability data is essential for clinical decision making and improved patient outcomes, yet standardized approaches to the analysis and visualization of tolerability data in cancer clinical trials are currently limited. To address this need, the Standardization Working Group (SWG) was established within the NCI Cancer Moonshot Tolerability Consortium. This abstract describes the SWG’s initiative to develop a publicly accessible online toolkit with a comprehensive set of guidelines, references, and resources for graphical displays of tolerability data. Methods: A multidisciplinary group of PRO researchers including biostatisticians, clinicians, epidemiologists, and representatives from the NCI and FDA convened monthly to discuss toolkit development and content. Considerations for standardization of graphical displays of tolerability data included (1) types of graphical displays, (2) incorporation of missing data, (3) labeling and color schemes, and (4) software to produce graphical displays. For consistency, considerations of tolerability relied on the Patient-Reported Outcomes version of the CTCAE (PRO-CTCAE), which includes 124 items assessing the frequency, severity, interference, and/or presence of 78 symptomatic adverse events. Graphical displays were generated using simulated PRO-CTCAE data and summarized by composite score (range 0-3).Color schemes that were Section 508 compliant and color blindness accessible were created. Surveys were distributed to 68 consortium members to assess preferences and interpretability of the graphical displays. Results: The SWG created graphical displays for PRO-CTCAE data, including bar charts, butterfly plots, and Sankey diagrams and compiled SAS macros and R functions to do so. Graphical displays made available in the toolkit maximize the use of PRO-CTCAE data, incorporate missingness, support between-arm comparisons, and present data longitudinally over treatment cycles or study timepoints. Survey results for labeling and color schemes were summarized and informed a list of short labels for PRO-CTCAE items (e.g., “radiation burns” for “skin burns from radiation”) and standardized color schemes for use in graphical displays. Survey results were also summarized to provide insight into PRO researchers’ ability to accurately interpret the graphical displays. Conclusions: Standardizinggraphical displays is important for improving the communication and interpretation of tolerability data. The type of graphical display used depends on the purpose of the analysis and should be tailored to the intended audience, including patients. This toolkit will provide a comprehensive resource with best practice recommendations.