Background: The Milan criteria (MC) are current standard for indicating liver transplantation (LT) for hepatocellular carcinoma (HCC). However, many beyond MC patients may be rejected despite beneficial outcome. Routine serological parameters of tumor viability, such as alpha-fetoprotein (AFP) and C-reactive protein (CRP), were shown to describe tumor aggressiveness. However, they have not yet been implemented in patients’ selection process. The aim of this study was to assess the prognostic accuracy of combining AFP and CRP to predict posttransplant outcome beyond standard criteria.
Patients and methods: 119 liver transplant patients with HCC were retrospectively analyzed. Tumors were clinically staged according the MC and the Up-to-seven (UTS) criteria. The most optimal cut-off values of AFP (100 ng/ml; Area under the curve [AUC] = 0.826; r = 0.042; 95%Confidence Interval [CI] 0.743 – 0.909) and CRP (0.8 mg/dl; AUC = 0.824; r = 0.039; 95%CI 0.747 – 0.901) for predicting tumor recurrence were determined by ROC analysis. The impact of pretransplant available radiographic and serological features on posttransplant outcome was analyzed by uni- and multivariate analysis.
Results: Median post-LT follow-up was 74 months (range: 5-184). Tumor recurrence rate was 24.4%. In multivariate analysis, only CRP < 0.8 mg/dl (Odds ratio [OR] = 36.2; p < 0.001) and AFP < 100 ng/ml (OR = 17.571; p < 0.001) were identified as independent prognosticators of HCC recurrence, whereas macromorphology features were not independently significant. Five year RFS rates were 97.7% in low CRP/AFP (n = 44) and 85.8% in low CRP or AFP (n = 50) patients, defining a low serological tumor viability risk index (STVI). In contrast, RFS was only 10% in high CRP/AFP patients (n = 25; high STVI; log rank < 0.001). RFS was not different between Milan In patients (n = 69; 86.8%) and Milan Out patients with low STVI (n = 32; 87.1%), and between UTS In patients (n = 88; 81.6%) and UTS Out patients with low STVI (n = 20; 89.2%). In contrast, it was 0% in both beyond Milan and beyond UTS patients yielding high STVI (log rank < 0.001), respectively. Application of STVI increased transplant eligibility between 23% and 46%.
Conclusion: We were able to demonstrate that a STVI implementing routinely determined AFP and CRP values selects suitable liver transplant patients beyond standard criteria. Patients with beyond Milan or UTS HCC and pre-LT elevated AFP and CRP levels should be rejected since they yield an extremely high recurrence risk.