A Model for Adaptive Decision Making of 'Ablate-and-Wait' Versus Transplantation in Patients with Hepatocellular Carcinoma

Abstract

Background/Aims: In patients with early-stage hepatocellular carcinoma (HCC), selection of candidates for liver transplantation (LT) requires refinement based on tumor biology to maximize the outcome. We aimed to prognosticate LT candidates with HCC using a risk prediction model for post-LT recurrence. Patients and Methods: A total of 197 consecutive patients were included who underwent LT for hepatitis B-related HCC within the Milan criteria. A risk prediction model was developed for post-LT recurrence using the Cox model and was internally validated. Results: Among those undergoing LT as their first HCC treatment (n=70, initial LT group), poor prognosis was associated with maximal tumor size and multinodularity. The remaining 127 patients (deferred LT group) received radiofrequency ablation (n=69) and/or transarterial chemoembolization (n=98) before LT. Multinodularity, maximal tumor size, posttransarterial chemoembolization progressive disease, baseline alpha-fetoprotein, and alpha-fetoprotein difference (between baseline and pre-LT) were incorporated into a risk prediction model for the deferred LT group, which was thereby stratified into low-risk (score<5), intermediate-risk, and high-risk (score?8) subgroups. Recurrence-free survival was significantly different among the deferred LT prognostic subgroups (P<0.001). Conclusions: This risk prediction model may help refinement of ablate-and-wait strategy for LT candidates by avoiding LT in those with either high risk score at baseline or increasing score under repeated locoregional therapies. © 2017 Wolters Kluwer Health, Inc. All rights reserved.