Comprehensive Molecular Analyses of a Six-Gene Signature for Predicting Late Recurrence of Hepatocellular Carcinoma.

Yuyuan Zhang,Libo Wang,Long Liu,Zaoqu Liu,Xinwei Han,Zhen Li,Xin Li

doi:10.3389/fonc.2021.732447

Abstract

A larger number of patients with stages I–III hepatocellular carcinoma (HCC) experience late recurrence (LR) after surgery. We sought to develop a novel tool to stratify patients with different LR risk for tailoring decision-making for postoperative recurrence surveillance and therapy modalities. We retrospectively enrolled two independent public cohorts and 103 HCC tissues. Using LASSO logical analysis, a six-gene model was developed in the The Cancer Genome Atlas liver hepatocellular carcinoma (TCGA-LIHC) and independently validated in GSE76427. Further experimental validation using qRT-PCR assays was performed to ensure the robustness and clinical feasible of this signature. We developed a novel LR-related signature consisting of six genes. This signature was validated to be significantly associated with dismal recurrence-free survival in three cohorts TCGA-LIHC, GSE76427, and qPCR assays [HR: 2.007 (1.200–3.357), p = 0.008; HR: 2.171 (1.068, 4.412), p-value = 0.032; HR: 3.383 (2.100, 5.450), p-value <0.001]. More importantly, this signature displayed robust discrimination in predicting the LR risk, with AUCs being 0.73 (TCGA-LIHC), 0.93 (GSE76427), and 0.85 (in-house cohort). Furthermore, we deciphered the specific landscape of molecular alterations among patients in nonrecurrence (NR) and LR group to analyze the mechanism contributing to LR. For high-risk group, we also identified several potential drugs with specific sensitivity to high- and low-risk groups, which is vital to improve prognosis of LR-HCC after surgery. We discovered and experimentally validated a novel gene signature with powerful performance for identifying patients at high LR risk in stages I–III HCC.

Highlights

Hepatocellular carcinoma (HCC) is a highly malignant cancer with a poor prognosis [1]
The liver hepatocellular carcinoma (LIHC) transcriptome profiles with clinical data were obtained from The Cancer Genome Atlas (TCGA, https://portal.gdc.cancer.gov/) and the chip-array profiles with clinical data were downloaded from Gene Expression Omnibus (GEO, https://www.ncbi.nlm.nih. gov/geo/)
Mutation landscape in late recurrence (LR)- and NR-HCC patients was firstly demonstrated in our study

Summary

Introduction

Hepatocellular carcinoma (HCC) is a highly malignant cancer with a poor prognosis [1]. Relative to the time from surgery to initial recurrence, HCC recurrence is typically divided into early and late recurrence (LR), which has generally been defined using 2 years as a cutoff value [4]. Many patients who were alive and free of tumor recurrence at 2 years after curative liver resection of HCC do not take surveillance regularly, and may lose the chance to undergo curative treatment when symptoms develop. The LR is probably related to the evolution of the underlying chronic liver diseases, which was generally considered a de novo tumor with different biologic behaviors compared with early recurrence [6]. It is researched that the relationship between cirrhosis and LR has clinical “face validity,” as chronic hepatitis inflammation and fibrosis accelerated HCC development by generating a carcinogenic microenvironment in the liver, known as the “field effect” [7]. We hope to translate this knowledge into new biomarkers and targets in order to have an impact on decision-making of surveillance and treatment, and improve the clinical outcomes of HCC patients

Objectives

Methods

Results

Discussion

Conclusion