Endothelial angiopoietin-2 overexpression in explanted livers identifies subjects at higher risk of recurrence of hepatocellular carcinoma after liver transplantation

Abstract

Background: Though the precise criteria for accessing LT are consistently being applied, HCC recurrence (HCC-R_LT) still affects more than 15% of the patients. We analyzed the clinical, histopathological, and biological features of patients with HCC to identify the predictive factors associated with cancer recurrence and survival after LT.

Methods: We retrospectively analyzed 441 patients with HCC who underwent LT in our center. Overall, 70 (15.8%) of them developed HCC-R_LT. We matched them by age at transplant and etiology with 70 non-recurrent patients. A comparable cohort from the Liver Transplant Centre of Bologna served as validation. The clinical and biochemical characteristics and pre-LT criteria (Milan, Metroticket, Metroticket_AFP, and AFP model) were evaluated. Histological analysis and immunohistochemistry for angiopoietin-2 in the tumor and non-tumor tissue of explanted livers were performed. Patients' follow-up was until death, last clinical evaluation, or 31 December 2021. In patients with HCC-R_LT, the date of diagnosis of recurrence and anatomical site has been reported; if a biopsy of recurrence was available, histologic and immunohistochemical analyses were also performed.

Results: Patients were followed up for a mean period of 62.7 ± 54.7 months (median, 39 months). A higher risk of HCC-R_LT was evident for factors related indirectly (AFP) or directly (endothelial angiopoietin-2, microvascular invasion) to biological HCC aggressiveness. In multivariate analysis, only angiopoietin-2 expression was independently associated with recurrence. Extremely high levels of endothelial angiopoietin-2 expression were also found in hepatic recurrence and all different metastatic locations. In univariate analysis, MELD, Metroticket_AFP Score, Edmondson-Steiner grade, microvascular invasion, and endothelial angiopoietin-2 were significantly related to survival. In multivariate analysis, angiopoietin-2 expression, Metroticket_AFP score, and MELD (in both training and validation cohorts) independently predicted mortality. In time-dependent area under receiver operating characteristic curve analysis, the endothelial angiopoietin-2 expression had the highest specificity and sensitivity for recurrence (AUC 0.922, 95% CI 0.876-0.962, p < 0.0001).

Conclusions: Endothelial angiopoietin-2 expression is a powerful independent predictor of post-LT tumor recurrence and mortality, highlighting the fundamental role of tumor biology in defining the patients' prognosis after liver transplantation. The great advantage of endothelial angiopoietin-2 is that it is evaluable in HCC biopsy before LT and could drive a patient's priority on the waiting list.

Keywords: angiopoietin-2; hepatocellular carcinoma; immunocytochemistry; liver transplantation; neoangiogenesis; recurrence; survival.

Figure 1

Immunohistochemical analysis of angiopoietin-2 in endothelia of primary and recurrent HCCs after LT. Immunostaining for angiopoietin-2 (20×) shows marked cytoplasmic and vascular endothelial positivity both in primary HCCs at explant (A, C, E, G) and in the respective recurrent tumors (B: liver; D: peritoneum; F: lung; H: kidney). Endothelial and parenchymal angiopoietin-2 expression in hepatic and extra-hepatic recurrent HCCs is more marked than in the corresponding primary tumor. Arrows indicate representative endothelial localization (scale bar: 6.7 μm).

Figure 2

Receiver operating characteristic (ROC) analysis. Receiver operating characteristic (ROC) curves for the endothelial angiopoietin-2 and the different clinical scores for HCC recurrence (A) and survival (B) after liver transplantation in 140 matched transplanted HCC patients. angiopoietin-2 discriminating capacity was significantly higher than that of all clinical or pathologic scores for both recurrence (A) and survival (B).