Zhaohui Su, the Vice President of Biostatistics at Ontada, shared a post on LinkedIn:
“This paper introduces Quantitative Bias Analysis (QBA) as an important methodology for evaluating the reliability of external control arms constructed from real-world data (RWD) in clinical trials, especially for rare cancer populations where randomized controlled trials (RCT) are often impractical or unethical.
QBA encompasses techniques to model and quantify systematic errors—such as missing data, measurement error, and unmeasured confounding—that cannot be addressed by conventional statistical adjustments. The article illustrates QBA’s application using a case study comparing pralsetinib trial data to real-world outcomes for RET fusion-positive advanced non-small cell lung cancer. Methods like tipping point analysis and E-values are highlighted for assessing robustness against bias from missing or unknown confounders.
The authors emphasize that QBA strengthens the credibility of comparative efficacy estimates, guiding researchers and decision-makers in interpreting real-world evidence for regulatory and health technology assessment purposes.
Reference: Thorlund K, Duffield S, Popat S, Ramagopalan S, Gupta A, Hsu G, Arora P, Subbiah V. Quantitative bias analysis for external control arms using real-world data in clinical trials: a primer for clinical researchers. J Comp Eff Res. 2024 Mar;13(3):e230147. doi: 10.57264/cer-2023-0147. Epub 2024 Jan 11. PMID: 38205741; PMCID: PMC10945419.”
You can also read:
Zhaohui Su: Embracing Real-World Data to Power Rare Cancer Research
