Predicting waitlist dropout in hepatocellular carcinoma: a narrative review

Area of Science:

• Hepatology
• Transplantation Medicine
• Oncology

Background:

• Liver transplantation is the primary treatment for hepatocellular carcinoma (HCC).
• Organ scarcity creates challenges in allocating liver transplants to HCC patients.
• Predicting waitlist dropout is vital for equitable organ distribution.

Purpose of the Study:

• To review existing prediction models for HCC waitlist dropout.
• To summarize factors influencing HCC patient progression on the waitlist.
• To assess the current state of predictive modeling in HCC liver transplantation.

Main Methods:

• A narrative review of published articles up to December 25, 2023.
• Inclusion of studies utilizing statistical and machine learning (ML) models for dropout prediction.
• Exclusion of articles not focused on predictive modeling.

Main Results:

• Tumor size, alpha-fetoprotein (AFP) levels, MELD score, and response to locoregional therapy (LRT) are key predictors of dropout.
• Most existing models are statistical, with limited application of ML models.
• Despite numerous attempts, no models for HCC waitlist dropout prediction are currently implemented in clinical practice.

Conclusions:

• Accurate and reliable models are needed to improve HCC patient stratification for liver transplantation.
• Enhanced research methodology and robust databases are essential for model development.
• Clinical applicability of predictive models requires further validation and refinement.