Abstract:BACKGROUND & AIMS: Outcomes of liver transplantation for hepatocellular carcinoma (HCC) are determined by cancer-related and non-related events. Treatments for hepatitis C virus infection have reduced non-cancer events among patients receiving liver transplants, so reducing HCC-related death might be an actionable end point. We performed a competing-risk analysis to evaluate factors associated with survival of patients with HCC and developed a prognostic model based on features of HCC patients before liver transplantation. METHODS: We performed multivariable competing-risk regression analysis to identify factors associated with HCC-specific death of patients who underwent liver transplantation. The training set comprised 1018 patients who underwent liver transplantation for HCC from January 2000 through December 2013 at 3 tertiary centers in Italy. The validation set comprised 341 consecutive patients who underwent liver transplantation for HCC during the same period at the Liver Cancer Institute in Shanghai, China. We collected pretransplantation data on etiology of liver disease, number and size of tumors, patient level of alpha-fetoprotein (AFP), model for end-stage liver disease score, tumor stage, numbers and types of treatment, response to treatments, tumor grade, microvascular invasion, dates, and causes of death. Death was defined as HCC-specific when related to HCC recurrence after transplantation, disseminated extra- and/or intrahepatic tumor relapse and worsened liver function in presence of tumor spread. The cumulative incidence of death was segregated for hepatitis C virus status. RESULTS: In the competing-risk regression, the sum of tumor number and size and of log(10) level of AFP were significantly associated with HCC-specific death (P < .001), returning an average c-statistic of 0.780 (95% confidence interval, 0.763-0.798). Five-year cumulative incidence of non-HCC-related death was 8.6% in HCV-negative patients and 18.1% in HCV-positive patients. For patients with HCC to have a 70% chance of HCC-specific survival 5 years after transplantation, their level of AFP should be <200 ng/mL and the sum of number and size of tumors (in centimeters) should not exceed 7; if the level of AFP was 200-400 ng/mL, the sum of the number and size of tumors should be <= 5; if their level of AFP was 400-1000 ng/mL, the sum of the number and size of tumors should be <= 4. In the validation set, the model identified patients who survived 5 years after liver transplantation with 0.721 accuracy (95% confidence interval, 0.648%-0.793%). Our model, based on patients' level of AFP and HCC number and size, outperformed the Milan; University of California, San Francisco; Shanghai-Fudan; Up-to-7 criteria (P < .001); and AFP French model (P = .044) to predict which patients will survive for 5 years after liver transplantation. CONCLUSIONS: We developed a model based on level of AFP, tumor size, and tumor number, to determine risk of death from HCC-related factors after liver transplantation. This model might be used to select end points and refine selection criteria for liver transplantation for patients with HCC. To predict 5-year survival and risk of HCC-related death using an online calculator, please see www.hcc-olt-nietroticket.org/.

Improve the Prediction of Liver Transplant Mortality Based on Pre-Transplant Factors: A Multi-Center Study from China: Mortality Prediction of LT.

A Statistical Prediction Model for Survival after Kidney Transplantation from Deceased Donors

Predictive values of three indexes on short-term prognosis in patients with liver failure treated by artificial liver support system

A brief model evaluated outcomes after liver transplantation based on the matching of donor graft and recipient

Predicting short-term survival after liver transplantation on eight score systems: a national report from China Liver Transplant Registry

Predicting Short-term Survival after Liver Transplantation using Machine Learning

Long-term survival prediction for transjugular intrahepatic portosystemic shunt in severe cirrhotic ascites: assessment of ten prognostic models

Prediction of Survival after Liver Transplantation for Chronic Severe Hepatitis B Based on Preoperative Prognostic Scores: A Single Center'S Experience in China

Development of a Novel Prognostic Nomogram for High Model for End-Stage Liver Disease Score Recipients Following Deceased Donor Liver Transplantation

A novel prognostic model based on serum levels of total bilirubin and creatinine early after liver transplantation.

A Novel Nomogram for Prognosis Stratification in Salvage Liver Transplantation: a National-Wide Study with Propensity Score Matching Analysis in China.

Analysis of risk factors of early mortality after liver re-transplantation

Predicting Long-Term Allograft Survival in Liver Transplant Recipients

Online interpretable dynamic prediction models for clinically significant posthepatectomy liver failure based on machine learning algorithms: A retrospective cohort study

Machine Learning for Post-Liver Transplant Survival: Bridging the Gap for Long-term Outcomes through Temporal Variation Features

The lung allocation score and other available models lack predictive accuracy for post-lung transplant survival

A Transformer-Based Deep Learning Approach for Fairly Predicting Post-Liver Transplant Risk Factors

PB2409: CLINICAL PROGNOSTIC MODEL OF LIVER FAILURE AFTER ALLOGENEIC HAEMATOPOIETIC STEM CELL TRANSPLANTATION

Prediction of 1-year mortality after heart transplantation using machine learning approaches: A single-center study from China

Metroticket 2.0 Model for Analysis of Competing Risks of Death after Liver Transplantation for Hepatocellular Carcinoma