Our summary
Generating insights for rare subgroups can be a challenge and may require combining data from various sources. This is particularly true in precision oncology, where genomically-defined populations are of interest. For example, one approach is pooling real-world data (RWD) to form a comparator group for single-arm clinical trials in these rare subpopulations, enabling research insights and regulatory decision-making.
However, the best practices for pooling heterogenous RWD into a comparator group are undefined. RWD has unique considerations, such as differences in data quality and organization compared to clinical trial data. There is no comprehensive framework specifically for pooling RWD that addresses which studies should be merged and how to integrate them.
To fill this gap in guidance, researchers from Janssen and Flatiron Health developed a framework to address the methodological considerations of pooling RWD, specifically applied to creating a comparator group.
Why this matters
The framework presented in this study offers guidance to researchers and sponsors, assisting them in making informed decisions and planning effective clinical trials for rare genomic subpopulations. Additionally the publication has important implications for regulatory decision-making, potentially streamlining the approval process for novel precision medicine treatments. Furthermore, the study addresses specific challenges, including variations in data quality and organization, which are essential factors to consider when pooling RWD for research purposes.