'Living' Systematic Review and IPD Meta-Analysis to Evaluate the Contributions of Major and Ancillary Imaging Features on the Diagnostic Accuracy of LI-RADS.

Hepatocellular Carcinoma (HCC) is the most common type of primary cancer seen in the liver. Imaging tests such as CT, MRI, and contrast-enhanced Ultrasound (CEUS) are used to detect and diagnose HCC. The Liver Imaging Reporting and Data System (LI-RADS) helps radiologists to categorize liver lesions in patients with liver disease based on their relative risk for developing HCC using imaging features on CT, MRI and CEUS. LIRADS is comprised of a combination of 5 major and 26 ancillary imaging features to standardize the classification of liver lesions. Understanding the impact of individual imaging features on the diagnostic accuracy of LIRADS not only allows for better classification of lesions, but also helps optimize the LI-RADS systems and consequently improves the care of all patients at risk for HCC. Our team will combine data from patients whose findings have been published in different studies from centres all over the world. This technique is called an individual participant data (IPD) meta-analysis. This technique allows for detailed analysis of the impact of each imaging feature on the probability of HCC. The large pool of diverse participants allows for exploration the relative contributions of imaging features that are linked to HCC. The purpose of this study is to perform a living prospective systematic review and IPD meta-analysis to evaluate the contributions of individual major and ancillary features on the diagnostic accuracy of LI-RADS. This means we will gather individual patient data from published studies on an ongoing basis. Using this information, we will regularly update our analyses to determine which imaging features are most useful for diagnosing HCC. Based on the continually updated evidence, our Knowledge User Partners, the LI-RADS steering group, will update the LI-RADS systems in order to better diagnose HCC.