Target trial emulation in oncology: current use and future directions

Area of Science:

• Observational studies
• Causal inference
• Oncology research

Background:

• Target trial emulation (TTE) enhances causal inference from observational data.
• Knowledge gaps exist regarding TTE application, challenges, and insights in oncology.

Purpose of the Study:

• To systematically review TTE studies in oncology.
• To identify common data sources, endpoints, biases, and challenges.
• To assess TTE's role in calibrating randomized controlled trials (RCTs).

Main Methods:

• Systematic review of 90 TTE studies in cancer areas via PubMed and Embase.
• Analysis of data sources, primary endpoints, biases, and RCT calibration.
• Evaluation of statistical and estimate agreement for RCT calibration.

Main Results:

• Registry databases and overall survival (OS) were common.
• Significant immortal time bias (55.6%) and prevalent user bias (38.9%) were noted.
• Only 42.9% of TTE studies calibrating RCTs met agreement criteria.

Conclusions:

• Limited fit-for-purpose data and uncertainty in result concordance hinder oncology TTE.
• Promoting regulatory acceptance and improving data integration are crucial.
• Recommendations include enhancing data sources, standardizing endpoints like OS, and modernizing criteria.