Machine learning to predict waitlist dropout among liver transplant candidates with hepatocellular carcinoma
Allison Kwong1, Bilal Hameed2, Shareef Syed3
1Division of Gastroenterology and Hepatology, Stanford University, Stanford, USA.
Insights
Predicting waitlist dropout for liver transplant candidates with hepatocellular carcinoma (HCC) is now simpler. A new 5-variable model accurately forecasts dropout risk, aiding transplant prioritization for HCC patients.
Area of Science:
- Hepatology
- Transplant Surgery
- Oncology
Background:
- Accurate outcome prediction for liver transplant candidates with hepatocellular carcinoma (HCC) is difficult.
- A novel prediction model was developed to identify patients at risk of waitlist dropout.
Purpose of the Study:
- To develop and validate a predictive model for waitlist dropout in liver transplant candidates diagnosed with HCC.
- To improve patient prioritization for liver transplantation.
Main Methods:
- Utilized data from 18,920 adult liver transplant candidates in the US listed with HCC.
- Employed random forest and Cox proportional hazards models, analyzing 1,181 variables.
- Identified 5 key predictive variables: AFP, tumor size, bilirubin, INR, and ascites.
Main Results:
- A 5-variable model identified 12 predictive features for waitlist dropout.
- The final Cox model achieved a concordance statistic of 0.74 in the validation set.
- An online calculator is available for clinical application: http://hcclivercalc.cloudmedxhealth.com/.
Conclusions:
- A simple, interpretable 5-variable model effectively predicts 3-, 6-, and 12-month waitlist dropout in HCC patients.
- This model aids in prioritizing HCC patients based on their transplant urgency.
- The tool can enhance clinical decision-making for liver transplant allocation.
Background:
Accurate prediction of outcome among liver transplant candidates with hepatocellular carcinoma (HCC) remains challenging. We developed a prediction model for waitlist dropout among liver transplant candidates with HCC.
Methods:
The study included 18,920 adult liver transplant candidates in the United States listed with a diagnosis of HCC, with data provided by the Organ Procurement and Transplantation Network. The primary outcomes were 3-, 6-, and 12-month waitlist dropout, defined as removal from the liver transplant waitlist due to death or clinical deterioration.
Results:
Using 1,181 unique variables, the random forest model and Spearman's correlation analyses converged on 12 predictive features involving 5 variables, including AFP (maximum and average), largest tumor size (minimum, average, and most recent), bilirubin (minimum and average), INR (minimum and average), and ascites (maximum, average, and most recent). The final Cox proportional hazards model had a concordance statistic of 0.74 in the validation set. An online calculator was created for clinical use and can be found at: http://hcclivercalc.cloudmedxhealth.com/.
Conclusion:
In summary, a simple, interpretable 5-variable model predicted 3-, 6-, and 12-month waitlist dropout among patients with HCC. This prediction can be used to appropriately prioritize patients with HCC and their imminent need for transplant.
More Related Videos
08:15A Hepatocellular Cancer Patient-Derived Organoid Xenograft Model to Investigate Impact of Liver Regeneration on Tumor Growth
Published on: February 2, 2024
12:24A Three-Dimensional Spheroid Model to Investigate the Tumor-Stromal Interaction in Hepatocellular Carcinoma
Published on: September 30, 2021
