Mo1776 Quantitative Gene Expression of Liver Enriched Transcription Factors Could Be Used to Predict Tumor Recurrence After Curative Treatment of Hepatocellular Carcinoma

Abstract

Background: Hepatocellular carcinoma (HCC) is one of the most common cancers worldwide, and has a good prognosis only when is treated curatively. The development of new systems that can predict the likelihood of recurrence after curative treatment, including molecular prognostic factors, is much needed. The liver enriched transcription factors (LETFs) are critical in inducing and maintaining hepatic phenotype during liver organogenesis and their expression level in HCC could have prognostic implications. The aim of our work was to investigate the expression profile of liver enriched transcription factors FoxA2, HNF6 (ONECUT1), C/Ebp-alpha and HNF1-beta in hepatocelullar carcinoma specimens from liver resections or liver explants after liver transplantations, in comparison to non-tumoral liver tissue from the same patients. Methods: The study group included 22 patients, 12 with liver resection and 10 with liver transplantation for HCC with a mean follow-up after treatment of 20 months. Total RNA was isolated from tumoral and non-tumoral tissue fragments and gene expression has been quantified by qRT-PCR, using beta-actin as reference gene. A higher than 5-fold change in relative gene expression has been considered significant. A statistical model for prediction of tumor recurrence has been developped by logistic regression using quantitative RT-PCR data. The clinical utility of the model has been evaluated by the C-statistic (equivalent of the area under the ROC curve). Results: In 14 patients (63.6%) a higher than 5-fold change in relative gene expression has been detected. A significat up-regulation of the studied liver enriched transcription factors has been found in 4 patients (18.1%). A significat down regulation has been identified in 11 patients (50%) out of which HNF6 down regulation was detected in 4 patients (18.1%) and HNF1beta down regulation in 8 patients (36.3%). Only in one patient (4.5%) dramatical changes in LETFs expression has been detected FoxA2 and HNF6 were not detectable and C/Ebp-alpha and HNF1beta expression had a 100 fold decrease. The patient had a tumor recurrence in 4 months after the treatment. Based on qRT-PCR data a statistical model to predict recurence after curative treatment has been generated. The significant variables in the model were HNF6 and FoxA2 expression. The model has a C-statistic of 0.98 suggesting an excelent clinical utility. A cut-off level of 0.67 of the prognostic score had a 80% sensitivity and a 94.1% specificity in predicting recurrence of HCC after curative treatment. Conclusions: A significant change in LETF gene expression has been identified in 63.6% of patients curatively treated for hepatocellular carcinoma. A prognostic score using HNF6 and FoxA2 quantitative expression could be proposed for prospective validation.