Risk-Benefit Benchmarking in Liver Transplantation for Hepatocellular Carcinoma: a cohort study based on the CLTR and UNOS database

Background: Numerous transplant criteria and selective models have been developed to determine eligibility for liver transplantation (LT) and stratify prognosis in hepatocellular carcinoma (HCC) patients. However, no study has comprehensively compared these tools to assess their clinical benefit to patients. So, we aim to compare existing transplant criteria and models in 3 crucial dimensions: 5-year overall survival (OS), survival benefit and beneficiary (the number of patients within criteria).
Methods: In this comparative study, we included LT recipients with HCC from China Liver Transplant Registry (CLTR) and United Network for Organ Sharing (UNOS). OS was calculated using Kaplan-Meier method, and recurrence rate was assessed using competing-risk method. Cox proportional hazards regression was applied to assess the survival of HCC patients without LT from the Surveillance, Epidemiology, and End Results (SEER) database. Median survival benefit of LT for UNOS and CLTR cohorts was defined as the difference between the median OS with (based on actual follow-up) and without LT (based on the SEER model). For each transplant criteria and models, Risk-Benefit Benchmarking was calculated by product of following 3 relative values: 5-year OS, median survival benefit and beneficiary.
Results: A total of 15,717 LT recipients with HCC were included in this study: 11,520 from the UNOS cohort (April 2012 to June 2023) and 4,197 from the CLTR cohort (January 2015 to September 2023). 13 transplant criteria and models were compared in this study. Milan criteria (CLTR: N=1,661; UNOS: N=8,375) had near the highest 5-year OS (UNOS: 82.4%, CLTR: 78.7%) in the two cohorts, Hangzhou criteria (UNOS: N=10,979; CLTR: N=2,641) provided the largest number of HCC patients with LT eligibility and Dallas criteria showed the highest median survival benefit (whole cohort: 75.2 months). All the criteria and models had acceptable 5-year OS (>73%), comparable median survival benefit (>71 months) and restricted mean survival time (>30 months).
Conclusion: This study demonstrated that moderately expanding LT criteria for HCC was feasible, with recipients achieving satisfactory survival outcomes. Incorporating tumor biological markers into LT criteria could provide patients with a more favorable Risk–Benefit Benchmarking.