Urgent need for prognostic markers for hepatocellular carcinoma in the light of genomic instability and non-coding RNA signatures

Abstract

In this editorial, we comment on an original article by Duan et al. Despite advancements in the diagnosis and treatment of hepatocellular carcinoma (HCC), the identification of suitable prognostic factors remains challenging. In their paper, Duan et al identified long non-coding RNAs (LncRNAs) to quantify genomic instability (GI) by combining LncRNA expression and somatic mutation profiles. They confirmed that the GI-derived LncRNA signature (GI-LncSig) could be an independent prognostic factor with the area under the curve of 0.773. Furthermore, the authors stated that GI-LncSig may have a better predictive performance than TP53 mutation status alone. However, studies exploring genetic markers for predicting the prognosis of HCC are crucial for identifying therapeutic targets and enhancing diagnostic and treatment strategies to mitigate the global burden of liver cancer.

Key Words: Genomic instability; Long non-coding RNA; RNA signatures; Hepatocellular carcinoma; Liver cancer; Prognosis; Prognostic markers; Diagnosis; Precision medicine

Core Tip: Despite challenges in diagnosing and treating hepatocellular carcinoma (HCC), identifying suitable prognostic factors is crucial. Duan et al proposed a novel approach using long non-coding RNAs (LncRNAs) to quantify genomic instability (GI) in HCC. Their study demonstrated that the GI-derived LncRNA signature is an independent prognostic factor, outperforming TP53 mutation status alone with an area under the curve of 0.773. These findings underscore the importance of exploring genetic markers for predicting HCC prognosis, aiding in identifying therapeutic targets and improving diagnostic and treatment strategies to alleviate the global burden of this disease.

INTRODUCTION

Hepatocellular carcinoma (HCC) is the most common (> 80%) primary malignancy of the liver worldwide. It is also the sixth most common malignancy and the fourth most common cause of cancer mortality worldwide[1]. As the mortality rate of HCC is the highest among all malignant tumors, and the incidence has been increasing in recent years, despite dramatic improvements in the effectiveness of treatment with the introduction of immune checkpoint inhibitors, the objective response rate to therapy remains essentially unsatisfactory[2,3].

In line with this, the search for novel strategies to overcome therapy resistance and improve overall survival (OS) rate puts genomic instability (GI) under the focus of malignant tumor management[4].

Specific molecular markers can quantify GI and predict disease outcomes and prognosis. Song et al[5] were able to construct a gene GI signature that could help predict the OS of patients with histologically verified HCC. This breakthrough in GI signatures potentially could direct treatment for HCC patients[5]. The dysfunction of long non-coding RNAs (LncRNAs) was shown to be closely related to cancer development, including HCC[6].

However, although many LncRNAs have been demonstrated to be related to GI, their clinical application as potential new predictive biomarkers in cancer has largely been unexplored. Knowing the genetic bases underlying the occurrence of liver cancer and describing them in detail can help define the role of high-risk genetic variations, such as somatic mutations and epigenetic changes[7]. Also by exploring the complex interplay between environmental factors and changes in genes, it was found that genetic screening can help identify high-risk patients, their early diagnosis through liquid biopsy, and advances in high-throughput sequencing technologies[7].

GI-DERIVED LNCRNASIGNATURE IN HCC PATIENTS

The original article by Duan et al[8], examined descriptively and in detail identification of GI-LncRNAs, establishment of the GI-derived LncRNA signature (GI-LncSig) and validation of the GI-LncSig in patients with HCC. The authors attempted to establish a GI-LncSig by combining two profiles: That of LncRNA expression and that of somatic mutation, again with the aim of determining the outcome of these patients[8].

Duan et al[8], based on the literature data of HCC epidemiology and promising data on LncRNAs as a potential prognostic factor for cancers, including HCC, are insufficient and evasive. The authors established a computational framework to identify LncRNAs that are directly related to GI, which, by combining LncRNA expression with tumor mutant phenotype, could be used as a biomarker to predict the clinical outcome of HCC patients[8].

It is well-known that currently, the therapy of liver cancer has yielded some benefits. However, available data indicates that the therapy methods currently employed in clinical practice are primarily ineffective. The objective effective rate of therapy remains essentially inadequate, and the majority of patients do not respond well. The 5-year OS rate of metastatic HCC remains suboptimal[9,10].

We discussed this original article due to the scientific and applied results that contribute to science and clinical practice. The authors obtained 88 GI-LncRNAs by comprehensively analyzing the LncRNA profile and somatic mutation downloaded from the TCGA database. Furthermore, five novel GI-LncRNAs were further investigated to specifically assess the clinical outcome of HCC patients. TP53 is a frequent mutation site in cancer, and its type is substantially related to a worse survival rate in HCC patients[11,12].

According to the GI-LncSig results obtained by Duan et al[8], high-risk individuals had a considerably greater TP53 mutation rate than low-risk patients. Furthermore, high-risk and low-risk individuals with TP53 mutations had significantly different survival rates. As a result, this is crucial for the customized prognosis of HCC patients[8]. One of the limitations was that the GI-LncSig was based on a single TCGA database, due to the limited availability of LncRNAs in the HCC samples in the GEO database. Therefore, further verification is required, using another independent, larger and more comprehensive database. Additionally, in-depth in vivo and/or in vitro studies are needed to verify the mechanisms underlying the occurrence and progression of HCC, as GI-LncSig was determined using the mutation hypothesis-based computational framework. However, the limitations of the study are not fatal, but are opportunities to inform future research.

There are other studies, such as those by Huang et al[13] and Wu et al[14], which also used similar methods to determine the outcome of this neoplastic disease.

However, Duan et al[8], in their study, randomly assigned all HCC patients to one of two groups: Training or testing. As a result, the computed prognosis-related LncRNAs differed, as did the established GI-LncSig formula. Furthermore, the AUC for the GI-LncSig in this trial was relatively high. The GI-LncSig performed well in both the independent testing and TCGA sets. Although this study evaluated HCC patients' GI indexes and constructed the GI-LncSig to assess patient outcomes, several limitations require additional investigation[8].

The presented original article also highlighted that the combination of LncRNA expression with mutant phenotype of the tumor, can be used as an unconventional biomarker for predicting the clinical outcome and to evaluate the prognosis of HCC patients[8]. All of these can direct clinical decision-making for these patients.

Consequently, further research should focus on extending therapy targets and identifying accurate biomarkers, which will aid in adjusting treatment options and avoiding the risks and costs associated with medication inefficiency and side effects. It is critical to discover new accurate biomarkers that can forecast the prognosis of HCC patients, change treatment plans, and reduce the risks and expenses associated with medication in terms of effectiveness and side effects.

CONCLUSION

In conclusion, the study by Duan et al[8] sheds light on the challenges in diagnosing and treating HCC while highlighting the potential of LncRNAs as prognostic indicators. Their identification of the GI-LncSig as an independent prognostic factor, with superior predictive performance compared to TP53 mutation status alone, underscores the importance of integrating molecular markers in HCC management. Further studies are needed to validate and refine the GI-LncSig signature across diverse patient cohorts and treatment settings. Additionally, investigations into the underlying mechanisms driving the predictive value of GI-LncSig, such as its role in tumor progression and response to therapies, would be beneficial. Moreover, incorporating multi-omics approaches and large-scale clinical trials can enhance our understanding of HCC heterogeneity and facilitate the development of personalized treatment strategies. These efforts are vital for improving patient outcomes and reducing the global burden of HCC.