Predictors of post-recurrence survival in hepatocellular carcinoma patients following liver transplantation: Systematic review and meta-analysis

Data on predictors of post-recurrence survival (PRS) of recurrent hepatocellular carcinoma (HCC) after liver transplantation (LT) have not been reviewed and analysed systematically. We aimed to systematically analyse all published data on the predictors for PRS.In accordance with PRISMA and MOOSE guidelines, online search of PubMed and EMBASE databases was done for all reports that evaluate the predictors of PRS based on multivariate analyses. Cumulative analyses of hazard ratios (HRs) and their corresponding 95% CIs were conducted to assess the potential predictors of PRS.Twenty-three studies met the inclusion criteria. Among the 11,868 patients involved, 1921 (16%) had HCC recurrence within a median time of 16 months. The following were recurrence and tumour-related predictors: time to recurrence (<1 year; HR: 1.97; p < 0.001), AFP level at recurrence(≥100 ng/ml; HR: 1.82; p < 0.001), multiple recurrence (HR: 1.22; p < 0.001), bone recurrence (HR: 2.10; p < 0.001), poor differentiation (HR: 1.52; p < 0.001), intrahepatic recurrence (HR: 0.91; p = 0.03), extrahepatic recurrence (HR: 1.87; p < 0.001), Milan criteria at LT (HR: 1.34; p < 0.001), microvascular invasion (HR: 1.59; p < 0.001), multiorgan recurrence (HR: 1.28; p < 0.001), and recurrent HCV infection (HR: 1.21; p < 0.001). The treatment-related predictors were as follows: surgical resection (HR: 0.33; p < 0.001), mTOR inhibitors (HR: 0.63; p < 0.001), sorafenib (HR: 1.00; p = 0.01), palliative treatment (HR: 3.07; p < 0.001), RFA (HR: 0.47; p < 0.001), and radiotherapy (HR: 1.19; p < 0.001).Systematic evaluation of these predictors could guide surgeons to design risk-adapted algorithms for the management of post-LT HCC recurrence to construct reliable predictive models and to design future prospective studies or clinical trials.