BACKGROUND:

The Oncology Quality, Characterization, and Assessment of Real-world Data (QCARD) Initiative, led by the FDA Oncology Center of Excellence, aims to improve the evaluation of RWD for regulatory decision-making. This new initiative defines key quality dimensions—relevance, reliability, and external validity—to assess the fit-for-use of RWD in oncology. Ontada's ON.Genuity platform, which integrates EHR data from more than 500 community oncology clinics within both The US Oncology Network and non-Network, as well as supplementary data sources, is implementing the QCARD framework to improve the quality and applicability of oncology RWD. This study presents the approaches and results of applying QCARD to ON.Genuity data specifically in the context of hematologic malignancies.

METHODS:

We assessed ON.Genuity data with QCARD principles by applying automated checks for reliability and integrating standardized sources to enhance completeness and overall fit-for-use. EHR data sources include structured and unstructured data collection, natural language processing (NLP), and human abstraction. Standardization was achieved using Fast Healthcare Interoperability Resources (FHIR) and Minimal Common Oncology Data Elements (mCODE). Metadata and lineage documentation were used to support traceability and transparency. External validation was performed using the National Death Index (NDI) and published data from the Surveillance, Epidemiology, and End Results (SEER), an authoritative source for cancer statistics in the United States.

RESULTS:

The analysis covered about 1 million cancer patients over the past decade, using more than 250 standardized variables across 64 cancer types, including 23 hematologic malignancies and over 20 clinical domains. Conformance was attained by standardizing ON.Genuity data using FHIR and mCODE standards. Structured EHR data included comprehensive patient characteristics, cancer diagnoses, medication orders and administrations, laboratory results, and various other observations, although some in-patient medication and outcome information was not available. Chart abstraction significantly enhanced data reliability. For example, in the multiple myeloma cohort, chart-abstracted data showed over 90% completeness for key variables such as diagnosis dates, treatment regimens including cell therapies, comorbidities, and ECOG status. The top five plausibility checks had a 99.9% pass rate in all domains. Mortality data in EHR were consistent with the NDI, showing identical median overall survival (p=0.9). ON.Genuity's cancer incidence rates over time were consistent with trends in SEER, reinforcing external validity. These assessments align with the European Medicines Agency (EMA)'s Data Quality Framework, which provides structured, actionable guidance for assessing the relevance, reliability, coherence, and completeness of RWD in regulatory decision-making, and confirm ON.Genuity's readiness for regulatory-grade real-world evidence (RWE) generation.

CONCLUSIONS:

The implementation of the QCARD within the ON.Genuity platform introduces a process for generating high-quality RWD and RWE in oncology. The fit-for-use of RWD should be assessed for each study, ensuring relevance, reliability, and external validity aligned with the research objectives. This research contributes to streamlining regulatory submissions and promoting the use of RWD in clinical development, health technology assessments, and post-marketing surveillance. The described implementation practices may serve as a reference for other therapeutic areas, emphasizing the importance of data quality in oncology research and evidence-based decision-making.

This content is only available as a PDF.
2025
Sign in via your Institution