Overview
A persistent challenge in oncology RWE is the lack of standardized methods to assess disease progression and treatment response in EHR-derived data. Unlike clinical trials, where disease burden is measured through structured imaging assessments according to a trial protocol, real-world assessment is done based on clinical judgment of the treating clinician. This study tests an approach adapted from solid tumors to use routine clinical documentation to define real-world progression (rwP) and real-world insufficient response (rwIR) variables and capture disease worsening and refractory disease events in patients with non-Hodgkin lymphoma (NHL).
The study tested these measures in over 6,000 patients with NHL and found they were both reliable and useful for understanding treatment outcomes, validating the NHL-specific rwP/rwIR approach as feasible, reliable, and scalable. Specifically, researchers assessed the performance through inter-abstractor agreement on critical details, observing strong correlations between rwP/rwIR-based endpoints, such as real-world progression-free survival (rwPFS) and event-free survival (rwEFS), with overall survival (rwOS).
Why this matters
The ability to systematically capture these events from EHR data represents a major advancement in RWE methodology, allowing for a more robust understanding of the outcomes of patients with NHLs treated with routine clinical care and generating insights that can support regulatory submissions, comparative effectiveness research, and clinical decision-making. By adapting a previously validated approach used in solid tumors, the study demonstrates that these variables can be feasibly, reliably, and scalably extracted from EHR data.