This is the final post in the Real-World Evidence (RWE) blog series where we will discuss the Friends of Cancer Research (Friends) whitepaper titled: “Recommendations for Use of Real-world Evidence in Oncology: Lessons Learned from the Friends of Cancer Research Real-world Evidence Framework. ”
The whitepaper, informed by the pilot projects we reviewed in our previous posts to identify implications of dataset and patient characteristics on real-world outcomes, outlines a process for assembling fit-for-purpose datasets and a common real-world protocol and provides recommendations for developing an RWE framework.
The RWE pilot projects set out to investigate how different elements of real-world data (RWD) can be leveraged to support drug development and add to the collective evidence supporting the use of RWE in oncology research. RWD can be used to generate evidence that reflects a larger and more representative patient population than is included in clinical trials and address timely clinical questions, including long-term safety of therapies. The whitepaper is a compilation of considerations accumulated from the five RWE pilot projects and leveraged to guide future investigations and inform regulatory and payer decisions. Based upon results from the RWE pilot projects, the whitepaper presents considerations for standardizing a common set of definitions and data elements, methodology for interpreting datapoints, and the process used to develop an RWE framework.
Establishing a core set of data elements to collect and standardize definitions enables greater comparability across RWE studies independent of data sources. Important considerations when creating a core set of data elements include identifying analytic variables that require a high level of harmonization, aligning on harmonized definitions, and reviewing the distribution of pre-specified variables as an internal check. Harmonizing definitions can reduce variability that may naturally exist between datasets given varying data sources and patient populations. These elements promote data quality and reduces missing data and variance within a study.
In addition to a core set of data elements, aligning on statistical methodology is crucial for calculations and interpretations of endpoints. Considerations for methodology include, understanding the source data of the endpoint information, determining appropriate endpoints, providing transparency on endpoint derivation, and assessing fit-for purpose to increase confidence in the endpoint. Applying appropriate analysis methods is an important consideration as information derived from RWE will depend on the methodology and definitions used to select the specific patient populations.
Finally, based on the RWE pilots, the whitepaper developed a process for assembling fit-for-purpose datasets and a protocol to guide future real world. This framework could potentially maximize the quality of analyses of RWD and evidence generation to support oncology research. Adjustments will be required to expand the framework beyond the oncology field, but the process outlined in the whitepaper to establish the fit-for-purpose model will be relevant when applying the framework in other settings. With widespread application, this framework will contribute toward a robust understanding of RWE and its application in drug development.
Pilot Project 1.0 and 2.0 of the Real-World Evidence projects, involving multiple data partners, international populations, and oncology disease settings, would not have been possible without the multi-stakeholder collaboration. This alliance was necessary to create thorough recommendations for researchers to grasp the full potential of RWD in clinical research, drug development, and regulatory processes.