Background & aims: Patients with T2 hepatocellular carcinoma (HCC) can obtain an exception that allows them to undergo liver transplantation with much lower actual Model for End-Stage Liver Disease (MELD) scores than patients without HCC. We compared patients who received liver transplants, with and without HCC, with regard to transplantation-related survival benefit.
Methods: We modeled the post-transplantation survival of adult, first-time liver transplant recipients with HCC (n = 9135) or without (n = 25,890) from 2002 through 2013 using Cox proportional hazards regression. We modeled waitlist survival of patients listed for transplantation with HCC (n = 15,605) or without (n = 85,229) using competing risks analysis and combined outcomes of death or liver failure (defined as MELD score ≥30). We used these survival models to calculate monthly transition probabilities and 5-year life expectancies. Survival benefit was calculated as the difference between post-transplantation and waitlist life expectancy.
Results: The 5-year survival benefit increased with actual MELD score for patients with and without HCC, ranging from just a few months in patients with low MELD scores (ie, 6-8) to 4 years in patients with the highest MELD scores (ie, 36-40). The survival benefit of patients with HCC was similar to that of patients without HCC who had the same actual MELD score, irrespective of tumor burden or serum level of α-fetoprotein. However, because patients with HCC received liver transplants when they had a lower mean MELD score (13.3 ± 6.2) than patients without HCC (21.8 ± 8.0), a much lower mean 5-year survival benefit was achieved by providing liver transplants to patients with HCC (0.12 years/patient) than patients without HCC (1.47 years/patient).
Conclusions: The HCC MELD exception policy has unintentionally resulted in a large reduction in transplantation-related survival benefit.
Keywords: Exception Points; Liver Cancer; Priority; UNOS.
Copyright © 2015 AGA Institute. Published by Elsevier Inc. All rights reserved.