Related Experiment Video
Updated: Jun 12, 2026
A Three-Dimensional Spheroid Model to Investigate the Tumor-Stromal Interaction in Hepatocellular Carcinoma
Published on: September 30, 2021
Trajectory Modelling of Prognostic Biomarkers Linked to Liver Cancer Risk: A Systematic Review
Bashayr Aldawsari1, Laura J Bonnett1, Maria Sudell1
1Department of Health Data Science, Institute of Population Health, University of Liverpool, Liverpool, UK.
Introduction:
Liver cancer ranks as the third leading cause of cancer-related mortality. We review studies that used longitudinal clustering methods to identify trajectories of risk factors associated with risk of developing liver cancer.
Methods:
A systematic search of articles was conducted across three databases (MEDLINE, PubMed, and Scopus). Articles using any longitudinal clustering methods to identify trajectories of risk factors linked to liver cancer risk were considered. The review is registered on PROSPERO (CRD 42023406334).
Results:
Twenty-four studies were included. The people at higher risk of developing liver cancer in the general population are individuals with increasing body mass index, increasing metabolic syndrome, higher C-reactive protein, low or rare physical activity, high fasting blood glucose, and high smoking rate. Individuals with elevated risk of liver cancer in populations with liver disease had higher alcohol consumption, De Ritis ratio, AFP, FIB-4, liver fibrosis score, and stable HBsAg. Patients who already had liver cancer with increasing AFP and decreasing HBV DNA had lower survival probabilities. Three longitudinal clustering methods were identified: latent class growth modelling, Gaussian mixture models, and joint latent class modelling.
Conclusion:
This is the first systematic review of longitudinal clustering methods for prognostic risk factors linked to liver cancer risk. The study highlights the utility of longitudinal data in predicting liver cancer outcomes.