Int. J. Biol. Sci. 2020, Vol.

16 1551

Ivyspring
International Publisher
International Journal of Biological Sciences
2020; 16(9): 1551-1562. doi: 10.7150/ijbs.44024
Review

Circulating tumor DNA as an emerging liquid biopsy

biomarker for early diagnosis and therapeutic
monitoring in hepatocellular carcinoma
Xiaolin Wu1, Jiahui Li1, Asmae Gassa2, Denise Buchner1, Hakan Alakus1, Qiongzhu Dong3, Ning Ren4,
Ming Liu5, Margarete Odenthal6, Dirk Stippel1, Christiane Bruns1, Yue Zhao1,7, and Roger Wahba1
1. Department of General, Visceral, Cancer and Transplantation Surgery, University Hospital of Cologne, Kerpener Straße 62, 50937, Cologne, Germany.
2. Department of Cardiothoracic Surgery, Heart Center, University Hospital of Cologne, Germany, Kerpener Straße 62, 5.937 Cologne, Germany.
3. Department of General Surgery, Huashan Hospital & Cancer Metastasis Institute & Institutes of Biomedical Sciences, Fudan University, 200032, Shanghai,
P.R. China.
4. Liver Cancer Institute & Zhongshan Hospital; Department of Surgery, Institute of Fudan-Minhang Academic Health System, Minhang Branch, Zhongshan
Hospital, Fudan University, 200032, Shanghai, P.R. China.
5. Affiliated Cancer Hospital & Institute of Guangzhou Medical University; Key Laboratory of Protein Modification and Degradation, School of Basic Medical
Sciences, 510095, Guangzhou, P.R. China.
6. Institute of Pathology, University Hospital of Cologne, 50937, Cologne, Germany.
7. Department of General, Visceral und Vascular Surgery, Otto-von-Guericke University, 39120, Magdeburg, Germany.

 Corresponding authors: Roger Wahba (roger.wahba@uk-koeln.de; Tel: +49-221-478-4803); Yue Zhao (yue.zhao @uk-koeln.de; Tel: +49-221-478-30601).

© The author(s). This is an open access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/).
See http://ivyspring.com/terms for full terms and conditions.

Received: 2020.01.16; Accepted: 2020.02.03; Published: 2020.03.05

Abstract
As one of the most common malignant tumors worldwide, hepatocellular carcinoma (HCC) is known for
its poor prognosis due to diagnosis only in advanced stages. Nearly 50% of the patients with the first
diagnosis of HCC die within a year. Currently, the advancements in the integration of omics information
have begun to transform the clinical management of cancer patients. Molecular profiling for HCC patients
is in general obtained from resected tumor materials or biopsies. However, the resected tumor tissue is
limited and can only be obtained through surgery, so that dynamic monitoring of patients cannot be
performed. Compared to invasive procedures, circulating tumor DNA (ctDNA) has been proposed as an
alternative source to perform molecular profiling of tumor DNA in cancer patients. The detection of
abnormal forms of circulating cell-free DNA (cfDNA) that originate from cancer cells (ctDNA) provides
a novel tool for cancer detection and disease monitoring. This may also be an opportunity to optimize the
early diagnosis of HCC. In this review, we summarized the updated methods, materials, storage of
sampling, detection techniques for ctDNA and the comparison of the applications among different
biomarkers in HCC patients. In particular, we analyzed ctDNA studies dealing with copy number
variations, gene integrity, mutations (RAS, TERT, CTNNB1, TP53 and so on), DNA methylation
alterations (DBX2, THY1, TGR5 and so on) for the potential utility of ctDNA in the diagnosis and
management of HCC. The biological functions and correlated signaling pathways of ctDNA associated
genes (including MAPK/RAS pathway, p53 signaling pathway and Wnt-β catenin pathway) are also
discussed and highlighted. Thus, exploration of ctDNA/cfDNA as potential biomarkers may provide a
great opportunity in future liquid biopsy applications for HCC.
Key words: circulating tumor DNA; cell-free DNA; liquid biopsy; biomarker, hepatocellular carcinoma, liver
cancer

Introduction
Hepatocellular carcinoma (HCC) is one of the therapeutic options: liver transplantation, liver
most lethal cancers worldwide with progressive resection or microwave ablation, in addition, survival
accumulation and poor prognosis. Early diagnosis is could also be prolonged by transarterial
crucial in HCC because it provides multiple curative chemoembolization (TACE), systemic therapy with

http://www.ijbs.com
Int. J. Biol. Sci. 2020, Vol. 16 1552

tyrosine kinase inhibitor and selective internal challenge in the development of liquid biopsy
radiation therapy (SIRT) [1]. Patients with different applications and should be discussed in this review.
stages of HCC show huge differences in prognosis. At
early stage(I) patients with HCC compared to patients Methods and materials
in advanced stage (III) shows a significant improved A systematical search in PubMed for the terms
5-year survival rate with 59% compared to 29% [2]. “liquid biopsy”, “cell-free DNA”, “circulating tumor
Unfortunately, HCC is asymptomatic at an early DNA” in combination with “liver cancer”,
stage, and the majority of HCC is detected in the “hepatocellular carcinoma”, “biomarkers”, including
palliative stage. Therefore, the early diagnosis of HCC all papers in English from 2013-2019 has been
can only rely on modern medical technology. At performed. To summarize the latest developments,
present, the clinical practice includes radiological we elaborate on an overview of ctDNA in HCC with
screening and monitoring for patients with defined aspects about storage, detection, clinical application,
risk factors (liver cirrhosis, viral or chronic hepatitis, biomarkers, gene alterations and the function of the
NAFLD, etc.) in combination with AFP measurement. alternated genes.
AFP is one of the most widely used tumor markers for
HCC. With a low sensitivity of 62.4% and a cut-off Storage and detection methods for cfDNA
value of 20 ng/ml, AFP is not sensitive and accurate The most common source for the extraction of
enough for early detection and may reveal cfDNA is peripheral blood. CfDNA could also be
false-negative results [3,4]. Imaging techniques detected in other body fluids, including saliva, ascites,
including (CT, MRI or CEUS) had improved the pleural effusion, urine or stool. Currently, it is still
sensitivity from 66% to 82% and the specificity to widely discussed, if cfDNA should be extracted from
more than 90% merely for detecting nodules with at serum or plasma. Some studies discovered that the
least 1cm diameter [5]. level of cfDNA was higher in serum compared to
Liquid biopsy could be a future alternative plasma [12]; because a part of cfDNA is released
strategy. In cancer research, it has developed rapidly during blood cell lysis in the clotting process in tubes
as a diagnostic and monitoring tool, which can be before centrifugation, which leads to contamination
easily collected and analyzed in non-solid biological [13]. Thus, plasma is the preferable biological sample
tissue. The term “liquid biopsy” encompasses in most studies.
circulating tumor DNA (ctDNA)/cell-free DNA Due to the short half-life of cfDNA, specialized
(cfDNA), circulating tumor cells (CTCs), circulating approaches are required for storage and extraction.
miRNAs, and exosomes [6,7]. In this context, Plasma should be separated no more than 4-6 hours
ctDNA/cfDNA is one of the most frequently analyzed by centrifugation in EDTA tubes, which are the
objects. recommended blood collection tubes [14]. The plasma
First reported in human peripheral blood in 1948 sample can be isolated from the whole blood by
by Mandel and Metais [8], cfDNA is found as performing a 2000 x g centrifugation. However, when
double-stranded fragments of approximately 150 to the blood has to be stored more than 6h, specialized
200 base pairs in length [9], with a half-life of less than cfDNA collection tubes (BCT and CellSave tubes)
an hour. CfDNA, from myeloid and lymphoid with fixatives are better. They can help to keep the
apoptotic cells, shows low levels in healthy cfDNA quality up to 96h, including high copy
individuals (averagely 10 to 15 ng per milliliter [10]). numbers [15]. Temperature is also an essential factor
The concentration of cfDNA can rise in the blood in affecting the quality of cfDNA. Recently, a study for
cases of carcinoma, surgery, inflammation and tissue cfDNA preservation found out that the background of
damage. cfDNA in the blood of the EDTA tube has a rise delay
Circulating tumor DNA (ctDNA), which merely at 4°C [16]. Storage at low temperature could improve
refers to fragmented DNA, originates from tumor the quality of blood specimens: the recommended
cells itself. It represents a part of cfDNA, although storage conditions for samples are feezing at -80°C.
ctDNA has a substantial fluctuant proportion ranging Techniques for the detection of cfDNA require
from <0.1% to >90% in cfDNA [11], it is more specific. high sensitivity and specificity due to the low cfDNA
In general, cfDNA levels in the blood are elevated in concentration and the lower level of ctDNA in the
patients with carcinoma compared to healthy blood. Then cfDNA can be detected to identify several
individuals. With lots of ctDNA released into the common DNA based changes, including copy number
circulatory system by tumor cell apoptosis or necrosis, variations (CNV), DNA mutation, gene methylation,
the quantity of ctDNA could reflect tumor burden in and gene fusions, which is reflecting the status of
cancer patients. How to detect and analyze ctDNA DNA in the tumor cells. Different kits extract cfDNA
from the background of normal cfDNA is a big base on different methods, including silica-based

http://www.ijbs.com
Int. J. Biol. Sci. 2020, Vol. 16 1553

columns and magnetic beads. The principles of these treatment. As the fragmented DNA is released from
methods are based on the characteristics of cfDNA: carcinoma cells, ctDNA carries tumor-specific genetic
the binding affinity of cfDNA molecules could be information or epigenetic changes, which cannot be
enriched by silica column and magnetic enrichment is found in normal cfDNA. This helps to distinguish
for the negatively charged phosphate backbone of ctDNA from cfDNA.
cfDNA [17,18]. Right now the evidence in which The molecular changes that are useful to be
extraction methods are optimal is rare. The evaluation biomarkers of ctDNA must fulfill the following
of extraction efficiency and quality depends on the criteria:
kits. Kit extraction may exhibit different extraction • The biomarkers should be detected in tumor
efficacy even using the same method [17]. cells and plasma DNA simultaneously in the
The detection methods for cfDNA with different same individual;
sensitivity and nucleotide coverage can be
• The biomarkers must not be detected in cfDNA
summarized as BEAMing [beads, emulsion,
of healthy individuals.
amplification and magnetics] [19], droplet digital PCR
and sequencing methods (Sanger sequencing, next- Significant genetic and epigenetic information
generation sequencing [NGS]) [20]. The changing alterations in tumors include ctDNA level, CNVs,
rules of the sensitivity and nucleotide coverage in gene integrity, gene mutations, DNA methylation
these analyses technology are that the less sensitivity, alterations and gene fusions, which can be divided
the more extended nucleotide coverage, and vice into quantitative determination and qualitative
versa [11], which can be observed clearly in dd PCR alteration. Commonly altered genes may be different
and NGS techniques. even in the same patient and also cohort depending
Digital droplet PCR (dd PCR), the third on the type of genetic alteration. For example, most
generation of the PCR technique, is a high-throughput commonly mutant genes are TP53 and CTNNB1,
technique. It provides high precision and sensitivity while the most common amplifications are found in
(0.01% or less for allele frequencies in cfDNA) [21]. On MET and CCND1 [24].
the other hand, due to the high sensitivity, ddPCR is Currently, many studies in the last five years
unable to cover a large number of sequences. NGS can have investigated the specific gene changes from HCC
form a gene panel with many sequences from tumor tissues that can also be observed in ctDNA as
whole-exome sequencing to targeted sequencing of potential specific biomarkers: ctDNA level, CNVs,
cfDNA, which can optimize the limits of detection. In gene integrity, gene mutations and DNA methylation
the research on the same batch of samples with two alterations (Table 1). The majority of the data are
detection techniques, ddPCR and NGS have shown gathered from Asian cohorts.
different characteristics. With the ascendant CfDNA quantitative determination
sensitivity, ddPCR revealed a higher detection rate
than NGS (38% and 14%) in somatic mutations of CfDNA quantitative determination focuses on
cfDNA. It also took a precondition that objective cfDNA level and ctDNA copy numbers variation. The
mutation sites must be certainly known in advance. In increase of cfDNA level and copy number in tumor
other words, NGS is more suitable for primary gene patients are explained by the raised number of
profiling on multi-gene panels [22]. apoptotic and necrotic carcinoma cells. It has been
However, the sensitivity and specificity of the proven in liver cancer, oesophageal cancer, breast
NGS methods could be constrained by the error rate cancer, and pancreatic cancer [22,52,53]. Recently, Yan
of DNA polymerase and sequencing reactions. L et al. clearly showed that the cfDNA concentration
Molecular barcoding methods can help to tag the in plasma was significantly higher in HCC patients
original template cfDNA with unique nucleotide compared to patients with liver fibrosis or healthy
barcode and allow correct gene changes to be individuals [25]. And Xu H et al. also discovered a
accurately identified. In the original research, it is more significant CNV score in HCC patient group
necessary to make a rational choice for different than the chronic liver disease group [26].
methods according to different objectives [23]. Gene integrity
Different biomarkers in ctDNA of HCC Gene integrity analysis is also an efficient way to
Due to the extraordinary amount and variability distinguish ctDNA from the non-tumoral cfDNA in
of ctDNA in the cfDNA background, it is important to the plasma of HCC patients, focusing on the size of
detect and distinguish effective ctDNA from normal ctDNA molecules. Tumor-derived DNA molecules in
cfDNA accurately. This turns to be the challenge of plasma have a particular size profile. Jiang P et al. in
promoting this technology in clinical diagnosis and 2015 illustrated that the size of ctDNA molecules in
peripheral blood of HCC patients was smaller than

http://www.ijbs.com
Int. J. Biol. Sci. 2020, Vol. 16 1554

those from healthy individuals. Short DNA molecules signatures forming preferred plasma DNA end
may preferentially carry the tumor-associated gene coordinates. Thus ctDNA end coordinates could be
aberrations [29]. Furthermore, ctDNA did not served as hallmarks of DNA fragments from tumors
fragment randomly, because a class of ctDNA as well [31].

Table 1. Different biomarkers of ctDNA for HCC

Patients Controls Ethnicity Sample Sample vol. Biomarkers Biomarker Type Positive Rate (%) Supplement information Ref.
24 62 (HBV) China Plasma 1ml Concentration CfDNA level - - [25]
31 8 (CLD) China Plasma - CNVs & SNVs CNVs - - [26]
34 0 China Plasma - CNV CNVs - CNVs correlating to tumor [27]
burden
151 14 Korea Plasma 1.5ml VEGFA Amplication - - [28]
(HV) & &
CNV CNVs

90 67 (CLD) China Plasma 3-4.8 ml CfDNA size Integrity - - [29]

36 (LC)
32 (HV)
53 16 (OLT) China Plasma 1ml Plasma DNA integrity Integrity - ALU as primer [30]
- - China Plasma 4ml Preferred plasma DNA Integrity - - [31]
end coordinates
27 0 Asia, Plasma - RAS (KRAS & NRAS); Mutation 44.4 Evaluated for RAS mutational [32]
Europe, TERT Mutation 63.0 status by BEAMing firstly
USA TP53 Mutation 48.1
CTNNB1 Mutation 37.0
66 35 (LC) Italy Plasma 200ul TERT Mutation - - [33]
41 (CLD)
7 0 Europe Plasma 3-6ml TERT Mutation 86 The large tumor was>5 cm [34]
TP53 or metastatic HCC
23 0 CTNNB1 9 The small (largest tumor<5
TSC1 cm), nonmetastatic HCC
RB1
APOB
et al.
66 0 China Plasma 5ml TP53 Mutation 60 - [35]
CTNNTB1 15.7
AXIN1 14.3
ARID1A 14.3
51 10 (LC) UK & Italy Plasma - ARID1A Mutation 11.7 - [36]
CTNNB1 7.8
TP53 7.8
29 0 China Plasma 1.5-1.8ml TP53 Mutation 50 - [37]
ATM 39
ALK 36
NPM1 36
CSF1R 36
KIT 32
ERBB4 32
SMAD4 29
FBXW7 29
PTEN
33 0 China Plasma 5-6ml TP53 Mutation 52–84 - [38]
CTNNB1
AXIN1
JAK1
EPS15
CACNA2D4
206 0 USA Plasma 5-6ml TP53 Mutation & 0.49 (range 0.06 - median mutant allele [24]
EGFR Amplification 55.03%) frequency (% cfDNA)
MET
ARID1A
MYC
NF1
BRAF
ERBB2
26 0 USA Plasma - 5hmC Methylation - - [39]
25 90 (HV) China Plasma 2ml 5hmC Methylation 44 - [40]
1204 392 (CLD or LC) China Plasma 3-6ml 5hmC Methylation - validation set: area under [41]
958 (HV and BT) curve (AUC)=88.4%
29 32 (HV) USA Plasma 5ml multiple CpG sites Methylation 94.8 - [42]
8 (HBV)
36 38 (HV;LC; CLD) China Serum 2ml RGS10 Methylation 94 - [43]
ST8SIA6
RUNX2
VIM
51 186 (LC) France Plasma 3.5ml SEPT9 Methylation 94.1 Initial Study [44]

http://www.ijbs.com
Int. J. Biol. Sci. 2020, Vol. 16 1555

Patients Controls Ethnicity Sample Sample vol. Biomarkers Biomarker Type Positive Rate (%) Supplement information Ref.
47 103 (LC) Germany 85.1 Replication Study
66 43 (CLD) United Serum 1-2ml INK4A Methylation 65 - [45]
States
8 8 (HV) France Plasma 1ml VIM Methylation 2.3 - [46]
FBLN1 -
32 38 (HV) France Plasma 1ml VIM Methylation 1.48 Odds ratios [47]
FBLN1 0.89
22 16 (CLD) Thailand Plasma 1ml VIM Methylation 2.18
28 (HV) FBLN1 0.75
31 27 (HV) China Serum - DBX2 Methylation 88 - [48]
31 (HBV) THY1 85
160 88 (CLD) China Serum 400ul TGR5 Methylation 48 - [49]
45 (HV)
121 37 (CLD) China Serum 400ul MT1M Methylation 84 - [50]
31 (HV) MT1G Methylation 70
715 560 (HV) China Plasma 1.5ml BMPR1A, Methylation 85.7 Diagnostic Panel [51]
PSD, ARHGAP25, KLF3,
PLAC8,
ATXN1,
Chr 8:20,
Chr 6:170,
Chr 6:3,
ATAD2
1049 - China Plasma 1.5ml SH3PXD2A, C11orf9, Methylation - Prognostic prediction Panel
PPFIA1, SERPINB5,
Chr 17:78, NOTCH3,
GRHL2, TMEM8B
Different kinds of biomarkers have been used to detect ctDNA from normal cfDNA, including cfDNA level, DNA copy number, gene integrity, gene mutations, and DNA
methylation alterations. In the past 5 years of ctDNA biomarker research, DNA methylation has become a research hotspot, followed by genetic mutation.

General mutation genes (RAS, TERT, CTNNB1,

Gene mutation TP53, AXIN, ARID1A) detected in ctDNA with a high
The number of somatic mutation sites in ctDNA incidence rate, were also in the top genomic panel
reflects tumor burden in most HCC patients and which was determined to be significantly mutated
represents the gene information of the primary cancer genes (SMGs). In HCC nodules, the oncogenes-TERT,
biopsy as well. In European HCC patient cohorts with CTNNB1, KRAS and NRAS mutations were present
large carcinoma (>5cm) or metastasis, at least one in 39%, 27%, 1% and 1% respectively [54, 55]. Tumor
mutation could be detected in 86% of the cases suppressor genes TP53 (31%) and AXIN1 (8%) were
showing that the genetic information in ctDNA inactivated by mutation, as well as the chromatin
reflected the condition of the primary tumor [34]. Lim remodeling genes ARID1A (7%) [55].
HY et al. showed further details of gene mutation in These mutant genes in ctDNA are involved in
HCC patients: 4.4% mutation for RAS could be several major signaling pathways. They play a key
detected by beaming technology. TERT, TP53, and role in the tumorigenesis and progression of HCC
CTNNB1 mutation could be detected in peripheral (summarized in Fig. 1), including MAPK/RAS
blood of HCC patients with RAS mutations confirmed pathway, Telomere maintenance, p53 signaling
by Beaming technology before. The frequencies of pathway, Wnt-β catenin pathway, and SWI/SNF
detected mutations of ctDNA corresponded with the complex related pathway [56].
somatic mutations in the hepatocellular carcinoma The MAPK/RAS is a ubiquitous signaling
were 44% (RAS), 63.0% (TERT), 48.1% (TP53), and pathway in all eukaryotic cells, which contains a chain
37.0% (CTNNB1) respectively [32]. In Chinese cohorts of proteins. It is located downstream of many growth
with high HBV infection frequency, TP53 had the factors for signal transduction from the cell surface to
highest mutation rate (60.0%). Frequently mutated the DNA in the cell nucleus. Therefore, MAPK/RAS
genes were CTNNB1 (15.7%), AXIN1 (14.3%), and pathway plays a fundamental role in the processes of
ARID1A (14.3%), while TERT mutation was not cell proliferation and survival for many malignant
included in the target deep sequencing (TDS) in the tumors, including HCC, non-small cell lung cancer,
original design. In total, 38.6% of HCC patients renal cell carcinoma or melanoma [57]. RAS gene
carried a mutation gene, which could be a potential belongs to oncogenes including KRAS, NRAS, and
direction for discovering novel therapeutic targets HRAS, separately coding KRAS, NRAS, and HRAS
[35]. Simultaneously, the frequency of common gene proteins in the RAS family [58]. As a small
mutations in HCC may vary in different cohorts. For molecular-weight GTP-binding protein, the RAS
example, ARID1A was detected as the gene with the family is crucial for cell growth control. RAS is an
highest mutation rate in ctDNA of HCC patients in important molecular signal regulator within this
some European cohorts [36] instead of TP53. pathway. Once RAS is active, it interacts with some

http://www.ijbs.com
Int. J. Biol. Sci. 2020, Vol. 16 1556

effectors to influence Raf protein kinases. Then the enough to classify liver tumors with activating
mitogen-activated protein (MAP) kinase cascade is CTNNB1 mutation and inactivating Axin mutation
activated by Raf, causing ERK activation to control the into the “Wnt/β-catenin” group at the same time.
cell cycle. At the ends of human chromosomes, ARID1A is a key subunit in SWI/SNF complex
telomerase is the cap for protection and could (Switch/Sucrose Non-Fermentable) for chromatin-
lengthen cell longevity. Telomerase reverse remodeling, and the ARID1A gene also mutated
transcriptase (TERT; or hTERT) is a catalytic subunit frequently in liver carcinoma patients. SWI/SNF
of the telomerase complex and has dominant roles for complex contains two large subclasses: BAF and
telomerase activity. Telomerase is inactivated in most PBAF. The BAF subunit ARID1A may take the highest
somatic cells and it will be reactivated only in mutation rate among other SWI/SNF components. As
proliferative cells and most cancer cells, including a tumor suppressor gene, ARID1A has been
liver cancer cells. The TERT expression in cancer cell confirmed to restrain cell proliferation. On the other
survival can be upregulated in HCC, especially in US-Data show that the expression levels of ARID1A
patients with hepatitis C virus infection [54]. are negatively correlated with survival level in
Currently, TERT can be a gatekeeper in the whole patients with HCC. This may indicate a dual role of
process from chronic hepatitis to HCC. In liver the ARID1A gene in tumorigenicity and cancer
cirrhotic tissues, TERT promoter mutation has been suppression for different temporal and cellular
able to identify in dysplastic nodules [59]. Tumor background in HCC [64].
protein p53 (TP53), also called p53, is classified as a
tumor suppressor gene. In humans, TP53 is an Gene methylation alterations
adapter in the DNA repair protein and can facilitate Mammalian DNA methylation alterations at the
damaged DNA repair by checkpoint arrest in the cell 5-position of cytosine (5 mC), a vital epigenetic
cycle. Mutations in the TP53 signaling pathway are regulator of gene expression profile occur in CpG sites
widespread in all kinds of tumors, including HCC. (cytosine-phosphate-guanine sites). It usually leads to
Common viruses and chemicals etiologic factors of alternations in DNA conformation, chromatin
HCC can prime TP53 mutations, including infection structure, and DNA protein interactions. DNA
with HBV and HCV, exposure particularly aflatoxin B methylation causes gene silencing and genome
and chronic inflammation [60]. TP53 mutation causes imprinting. As DNA epigenetic modification that has
DNA damage repair to be hindered, leading tumor been linked to DNA regulation and carcinoma
suppression is compromised and “gain function” in pathogenesis, it could appear in the early stage of
HCC cells, including uncontrolled proliferation, drug tumor development. Therefore, DNA methylation is a
resistance, and cancer cell migration. TP53 mutation biomarker in various human cancer types. Currently,
of ctDNA has a high detection rate in HCC the elevation of methylation hallmarks from ctDNA in
notwithstanding that, the evidence for TP53 mutation the plasma is one of the most extensively investigated
in HCC ctDNA as tissue-specific is rare, because of the fields in liquid biopsy in HCC. In the last 5 years,
prevalence in multiple malignant tumors. The Wnt nearly 50% of the studies about biomarkers in HCC
signaling pathway is an important cascade that focus on gene methylation. Studies on methylation
contains a series of signal transducer factors for signal biomarkers in ctDNA can be divided into three
delivery. With the effect of stemness and categories: methylation sites quantity, methylation
development regulation, Wnt is one of the most site expression and 5-hydroxymethylcytosine (5 hmC)
frequently activated pathways associated with HCC detection.
progression [61]. The oncogene β-catenin (CTNNB1) A method, called ‘CancerDetector’ was used to
and the tumor suppressor Axin play a crucial role in probabilistically model the joint methylation states of
the Wnt signaling pathway. In canonical Wnt multiple adjacent CpG sites on an individual
signaling with gene transcription regulation, sequencing read, in order to exploit the pervasive
transcriptional co-activator CTNNB1 will help to nature of DNA methylation generated for signal
stimulate gene expression, causing cell proliferation, amplification. This could detect methylation sites
anti-apoptosis and angiogenesis [62]. Furthermore, sensitively and further identify cancer at an early
CTNNB1 can cooperate with TERT to induce liver cell stage [42].
transformation. The tumor suppressors gene Axin is a Besides the amount of multiple CpG and 5hmC,
scaffolding protein, which has been proved to be a a large number of independent hypermethylated gene
negative regulator of the Wnt pathway a long time sites in ctDNA from HCC patients were identified,
ago. However, there has been a study showed Axin such as DBX2, THY1, TGR5, MT1M, MT1G, INK4A,
mutation may be independent of Wnt/β-catenin VIM, FBLN1, RGS10, ST8SIA6, RUNX, and SEPT9.
pathway in some HCC paitents [63]. It is not accurate DBX2 and THY1 methylation sites were proved to be

http://www.ijbs.com
Int. J. Biol. Sci. 2020, Vol. 16 1557

Figure 1. Main mutation pathways and functions in HCC development. MAPK/RAS pathway(marked with yellow), TERT mutation(marked with blue), p53 signaling
pathway(marked with green), Wnt-β catenin pathway(marked with gray), and SWI/SNF complex related pathway(marked with light red) are the common centralized signaling
pathways. They affect tumorigenesis and progression of hepatocellular carcinoma, involving several common oncogenes and tumor suppressors. Corresponding functions
include proliferation, immortalization, genomic stability, cell differentiation and survival. The mutation genes detected in ctDNA show significant roles in pathways (Red box).

detectable in ctDNA by Infinium Human Methylation et al. gathered several methylation points from two
450 BeadChip in early-stage HCC patients [48]. Han gene panels: one was developed as a diagnostic panel
LY et al. discovered the value of DNA methylation at for HCC, which includes ten different methylation
G-protein-coupled bile acid receptor Gpbar1 (TGR5) points: BMPR1A, PSD, ARHGAP25, KLF3, PLAC8,
in ctDNA as a biomarker for HCC patients with ATXN1, Chr 6:170, Chr 6:3, ATAD2. The other could
chronic hepatitis B (CHB) [49]. MT1M and MT1G predict prognosis, including SH3PXD2A, C11orf9,
methylation were detected in HCC patients´ serum PPFIA1, Chr 17:78, SERPINB5, NOTCH3, GRHL2 and
with a significantly higher frequency compared to TMEM8B [51].
patients with CHB or the healthy control group [50]. Discriminated with methylation, 5-
Occurring early in liver tumor progression with methylcytosine (5mCs) are converted to 5hmCs by
higher levels in HCC patients, INK4A promoter 5mC hydroxylase TET1, causing DNA demethylation
hyper-methylation also showed the potential as a alternation in mammalian cells. Recent research found
biomarker [45]. out that 5hmC was decreased in HCC patients at a
VIM methylated gene repeatedly appeared as very early stage, which was mainly because of the
the research focus in several studies, manifesting its decrease of 5mC associated with the hepatitis virus
significance as a biomarker for ctDNA in HCC. Wen L and activity of the TET enzyme [65]. Therefore, 5hmC
et al. showed that four hypermethylated CpG islands might be a biomarker of ctDNA with outstanding
(RGS10, ST8SIA6, RUNX2 and VIM) could be detected value for human cancer diagnosis and staging,
in blood and HCC tissue; simultaneously from dozens including liver cancer, lung cancer, colorectal and
of high-performance markers of HCC by MCTA-Seq gastric cancer, and pancreatic cancer [39,40].
technique [43]. In another study, a panel of genes with The methylated genes originally have different
methylation of ctDNA was assessed in two different functions, including DNA damage, metabolic
cohorts (France and Thailand), discovered that regulation, apoptosis, G protein-coupled signal
changes in VIM and FBLN1 methylation levels could transmission, and some plasma protein release (Fig.
represent effective ctDNA biomarkers [47]. SEPT9 2). RUNX2 is a critical factor for osteoblast
was also verified to be the circulating epigenetic differentiation, interacting with the p53 tumor
biomarker in plasma for tumor diagnosis at an suppressor gene for DNA damage signaling cascade
individual level [44]. [66]. THY1 and ST8SIA6 are both involved in
In addition to sporadic individual genes, Xu RH post-translational protein modification. As a

http://www.ijbs.com
Int. J. Biol. Sci. 2020, Vol. 16 1558

glycophosphatidylinositol (GPI) anchored protein, increased in acute liver injury [75].

GPI-THY1 can regulate signals impacting cell
adhesion, differentiation and migration [67]. ST8SIA6 CtDNA combines with protein markers
is one of the sialyltransferases acting on N-linked Recently, there was an interesting study by
glycosylation for post-translational protein Cohen JD et al. for a new liquid biopsy method called
modification [68]. MT1M and MT1G, members of “CancerSEEK” for tumor early detection, which
metallothioneins (MTs), play vital roles in metal combined mutations in ctDNA and circulating
homeostasis, metal detoxification, and metabolism of proteins for eight common cancer types (ovary, liver,
vitamins and cofactors [69]. RGS10 is a stomach, pancreas, esophagus, colorectum, lung and
GTPase-activating protein. It is a specific type of Gα breast). In 1005 cancer patients the test was positive in
subunits and can attenuate the signaling pathway for a median of 70% (ranging from 69 to 98%), and
heterotrimeric G proteins. As a specific type of Gα specificity was more than 99% for the different tumor
subunits, RGS10 affects the production of cAMP, entities, compared with 812 healthy controls. The
further indirectly affecting fatty acid metabolism [70]. sensitivity of liver cancer was nearly 98%. It reaches
During apoptosis, vimentin (VIM) is cleaved by almost 100% in early tumor detection (for stage I HCC
several caspases, which can produce pro-apoptotic patients). Limitations of this study were the small size
amino-terminal production to amplify the cell death of a cohort including only 44 patients with liver cancer
signal [71]. Abundantly expressed in cell membranes (39 hepatocellular carcinoma and 5
of gallbladder epithelium, TGR5 (GPBAR1) is a cholangiocarcinomas). This study showed how
G-protein-coupled bile acid receptor for bile acid effective ctDNA could be applied for liver cancer
mediating [72]. Among the septin family, which is as detection [76].
GTP-binding proteins for cell division, SEPT9 is a key
Clinical application of ctDNA detection in
factor for cytokinesis and has linked alterations to
HCC patients in liquid biopsies
cancer development [73]. INK4A (CDKN2A), which
belongs to the family of cyclin-dependent kinase Future application of liquid biopsy represents a
inhibitors, leads to cell cycle arrest in the G1 phase by critical direction with broad clinical prospects.
causing the inhibition of cyclin D-dependent kinases CtDNA has many advantages: non-invasive, multiple
[74]. Fibulin-1(FBLN1) belongs to an enormous family time points monitoring, characterization of cancer,
of plasma glycoprotein and will be significantly identification of mechanisms of resistance. Compared

Figure 2. Functions of the genes with methylation. The functions of the genes with methylation (marked with red) mainly focus on several aspects: DNA damaged, metabolic
regulation, apoptosis, G protein-coupled signal transmission, cell division, and some plasma protein release. RUNX2 has an influence on DNA damage(marked with blue); THY1,
ST8SIA6, MT1M, MT1G, and RGS10 participate in the metabolism process (marked with green); VIM has a connection with apoptosis (marked with yellow); TGR5(GPBAR1) is
a G-protein-coupled bile acid receptor for bile acid mediating; SEPT9 plays a critical role of cytokinesis and INK4A (CDKN2A) is a cell cycle inhibitor(marked with light red);
FBLIN1 is related to plasma glycoprotein generation(marked with orange).

http://www.ijbs.com
Int. J. Biol. Sci. 2020, Vol. 16 1559

with standard tumor biopsy, blood samples are easy treatment after surgery and the timing of TACE
to obtain multiple times in a non-invasive method. treatment. Simultaneously, profiling the molecular
Thus, for patients who do not need or cannot undergo changes in ctDNA/cfDNA may be able to guide
surgery, the non-invasive ctDNA analysis provides an targeted therapy. Sorafenib, a multikinase inhibitor
option of molecular information. It allows a practical for anti-angiogenesis and anti-proliferation, can
way of monitoring tumor change in patients serially prolong nearly 3 months of median survival time for
over time without physical injury. advanced HCC patients. HCC patients receiving
At present, diagnosis at late-stage and tumor sorafenib treatment show the amplification of
recurrence remains to be a severe challenge in HCC VEGFA copy number [28]. CtDNA may help to
treatment, because of due limited techniques for monitor the efficacy of treatment and time to detect
clinical use. A tumor is always caused by a gradual drug resistance for HCC patients [81].
increase of genetic aberrations that manage cell CtDNA has also the potential for distinguishing
proliferation and apoptotic. Thus ctDNA detection tumor subtypes. This could be vital to guide diagnosis
using genetic changes as a detection indicator has the and therapy. Right now there are no relevant studies
huge potential to detect tumor cells at early stages. In for HCC ctDNA differentiation of subtypes. However,
the research of Xu RH et al, ctDNA methylation the ctDNA genotype has been proven to classify
markers showed different scores between early-stage tumor subtypes in diffuse large B cell lymphoma [82].
disease (I, II) and advanced-stage disease (III, IV) [77], In lung cancer, ctDNA may detect two different
reflecting the potential of ctDNA to be an early subtypes in the EGFR mutant gene [83]. Thus it is
diagnostic tool for HCC indirectly. It is also in necessary and urgent to explore the potential of
multiple tumors at early stages as well, including HCC-ctDNA regarding these aspects.
colorectal, ovarian, lung, and breast cancer [78]. Simultaneously, the exploration of genetic
Moreover, tumor cell residual is one of the important aberration in HCC ctDNA mainly depends on known
factors for postoperative tumor relapse. Owing to the genetic changes in tumor tissues at present. It has
short half-life of ctDNA, any changes in tumor- certain limitations: detection rates of mutated genes in
derived ctDNA are able to provide clear evidence of different populations in ctDNA did not match
real-time development of carcinoma, and this could detection rates in tumor tissues. The increasing
help to detect postoperative tumor residuals and development of artificial intelligence (AI), in medical
guide following treatment. As ctDNA correlates with imaging for cancer patients (e.g. lung, brain, breast,
tumor burden, it could be a tool for recurrence and prostate cancer [84]) could be combined with
monitoring for HCC. database of HCC tumor DNA, ctDNA and HCC
CtDNA is an emerging technique with immense marker proteins with AI could lead to exploring new
potential for different clinical applications in HCC biomarkers, establishing precision oncology
such as early tumor detection, therapy evaluation or detection, diagnosis, therapy and monitoring for
monitoring of metastasis. Therefore, ctDNA may patients.
reflect tumor heterogeneity and subclone mutation
during disease progression. Gene alteration could be Conclusions
made mutation or methylation panel to assess the In this review, we summarized the studies on
prognosis in gene profiling [51]. Precision oncology is biomarkers for the detection of ctDNA from a normal
defined as seeking targeted molecular profiling and cfDNA background in HCC patients in past the 5
therapy on carcinoma to improve the patient’s years: CNVs, gene integrity, gene mutations, DNA
prognosis. It aims to deliver the appropriate methylation, and cfDNA combined with protein
personalized cancer treatment to each individual in markers. In liver cancer research, the number of
terms of techniques, time and dose. The merits of studies on methylation ranks the highest, followed by
ctDNA which we have discussed above in this article studies about gene mutation. Rather than being
show the potential of its promising applications in limited to ctDNA quantity and size, exploring ctDNA
precision oncology [79]. from the perspective of alterations in DNA genetic
and epigenetic information has become a hot topic.
Future prospects These biomarkers offer new opportunities to improve
TACE is another first-line treatment for HCC the sensitivity and specificity of ctDNA analysis and
patients, aiming to induce tumor necrosis [80]. contribute to achieving ctDNA technology for clinical
Simultaneously, tumor necrosis is also one of the applications. However, the challenge for current
mechanisms of ctDNA release from the cell. Thus, in studies of ctDNA in HCC is the limitation of clearly
the future, the variation of ctDNA may be compelling defined useful genetic targets and biomarkers, which
data for whether to choose TACE as a supplementary can help to distinguish circulating tumor DNA from

http://www.ijbs.com
Int. J. Biol. Sci. 2020, Vol. 16 1560

normal cell-free DNA. 3. Trevisani F, D'Intino PE, Morselli-Labate AM, Mazzella G, Accogli E,
Caraceni P, et al. Serum a-fetoprotein for diagnosis of hepatocellular
CtDNA detection has still several limitations: carcinoma in patients with chronic liver disease: influence of HBsAg and
results of ctDNA analysis dependent on different anti-HCV status. Journal of Hepatology. 2001; 34: 570-5.
4. Okajima W, Komatsu S, Ichikawa D, Miyamae M, Ohashi T, Imamura T,
extraction sources, methods, sample size, blood et al. Liquid biopsy in patients with hepatocellular carcinoma:
collection time and conditions difference between Circulating tumor cells and cell-free nucleic acids. World J Gastroenterol.
2017; 23: 5650-68.
primary and recurrent tumors [85]. Standardization of 5. Villanueva A. Hepatocellular carcinoma. The New England Journal of
all these factors in different studies, as well as the Medicine. 2019; 380: 1450-62.
procedure for blood collection and storage, will be 6. Li J, Han X, Yu XN, Xu ZZ, Yang GS, Liu BQ, et al. Clinical applications
of liquid biopsy as prognostic and predictive biomarkers in
keys for guaranteeing consistency of results. Despite hepatocellular carcinoma: Circulating tumor cells and circulating tumor
the high incidence of liver cancer and promising DNA. Journal of Experimental & Clinical Cancer Research 2018; 37: 213.
7. Felden Jv, Craig AJ, Villanueva A. Role of circulating tumor DNA to help
results, the number of patients included in ctDNA decision-making in hepatocellular carcinoma. Oncoscience. 2018; 5 : 7-8.
clinical trials is still limited. Therefore, more 8. Mandel P, Metais P. Les acides nucléiques du plasma sanguin chez
l’homme. C R Seances Soc Biol Fil. 1948; 142.
researches are expected to take further exploration of 9. Fan HC, Blumenfeld YJ, Chitkara U, Hudgins L, Quake SR. Analysis of
more accurate biomarkers of ctDNA or different the size distributions of fetal and maternal cell-free DNA by paired-end
sequencing. Clinical Chemistry 2010; 56: 1279-86.
combinations of ctDNA with other effective markers.
10. Leon SA, Shapiro B, Sklaroff DM, Yaros MJ. Free DNA in the serum of
cancer patients and the effect of therapy. Cancer research 1977; 37:
Supplementary Material 646-50.
11. Corcoran RB, Chabner BA. Application of cell-free DNA analysis to
Supplementary appendix. cancer treatment. The New England Journal of Medicine. 2018; 379:
http://www.ijbs.com/v16p1551s1.pdf 1754-65.
12. Zinkova A, Brynychova I, Svacina A, Jirkovska M, Korabecna M.
Cell-free DNA from human plasma and serum differs in content of
Acknowledgements telomeric sequences and its ability to promote immune response.
Scientific Reports. 2017; 7: 2591.
We would like to thank the graphic resource 13. Martignano F. Cell-Free DNA: An Overview of Sample Types and
from Servier Medical Art. Some symbols in the Isolation Procedures. Methods Mol Biol. 2019; 1909: 13-27.
14. van Dessel LF, Beije N, Helmijr JC, Vitale SR, Kraan J, Look MP, et al.
manuscript and graphic abstract were modified from Application of circulating tumor DNA in prospective clinical oncology
https://smart.servier.com. trials - standardization of preanalytical conditions. Mol Oncol. 2017; 11:
295-304.
Author Contributions 15. Sherwood JL, Corcoran C, Brown H, Sharpe AD, Musilova M, Kohlmann
A. Optimised pre-analytical methods improve kras mutation detection in
Conceptualization, Y.Z, C.J.B, R.W and H.A; circulating tumour DNA (ctDNA) from patients with non-small cell lung
cancer (nsclc). PLoS One. 2016; 11: e0150197.
methodology, X.L.W, D.B and J.H.L; software, X.L.W; 16. Li SP, Bartlett B, Popoli M, Adleff V, Tucker L, Steinberg R, et al. The
validation, Q.Z.D; formal analysis, H.A, N.R; effect of preservative and temperature on the analysis of circulating
investigation, X.L.W; resources, Y.Z, R.W; data tumor DNA. Clinical Cancer Research. 2017; 23: 2471-7.
17. Sorber L, Zwaenepoel K, Deschoolmeester V, Roeyen G, Lardon F, Rolfo
curation, X.W; writing—original draft preparation, C, et al. A comparison of cell-free DNA isolation kits: Isolation and
X.W Y.Z; writing—review and editing, M.L, A.G, quantification of cell-free DNA in plasma. The Journal of Molecular
Diagnostics. 2017; 19: 162-8.
R.W; visualization, X.W; supervision, D.S, C.J.B; 18. Elazezy M, Joosse SA. Techniques of using circulating tumor DNA as a
project administration, Y.Z, R.W; funding acquisition, liquid biopsy component in cancer management. Computational and
Structural Biotechnology Journal 2018; 16: 370-8.
C.J.B, N.R, Y.Z, R.W. 19. Garcia-Foncillas J, Alba E, Aranda E, Diaz-Rubio E, Lopez-Lopez R,
Tabernero J, et al. Incorporating BEAMing technology as a liquid biopsy
Funding into clinical practice for the management of colorectal cancer patients: an
expert taskforce review. Annals of Oncology. 2017; 28: 2943-9.
This research was supported by the European 20. Shendure J, Ji H. Next-generation DNA sequencing. Nature
Fund for Regional Development (EFRE) for the Biotechnology. 2008; 26: 1135-45.
project “Autonomy in old age—potential biomarkers” 21. Hindson BJ, Ness KD, Masquelier DA, Belgrader P, Heredia NJ,
Makarewicz AJ, et al. High-throughput droplet digital PCR system for
and International Science and Technology absolute quantitation of DNA copy number. Analytical Chemistry. 2011;
Cooperation Project of Shanghai (18410721900); X.W. 83: 8604-10.
22. Pasternack H, Fassunke J, Plum PS, Chon SH, Hescheler DA, Gassa A, et
and J.L. were financially supported by CSC al. Somatic alterations in circulating cell-free DNA of oesophageal
scholarships (The China Scholarship Council). carcinoma patients during primary staging are indicative for
post-surgical tumour recurrence. Sci Rep. 2018; 8: 14941.
23. Newman AM, Bratman SV, To J, Wynne JF, Eclov NC, Modlin LA, et al.
Competing Interests An ultrasensitive method for quantitating circulating tumor DNA with
broad patient coverage. Nature Medicine. 2014; 20: 548-54.
The authors have declared that no competing 24. Kaseb AO, Sánchez NS, Sen S, Kelley RK, Tan B, Bocobo AG, et al.
interest exists. Molecular profiling of hepatocellular carcinoma using circulating
cell-free DNA. Clinical Cancer Research. 2019.
References 25. Yan LL, Chen YH, Zhou JY, Zhao H, Zhang HH, Wang GQ. Diagnostic
value of circulating cell-free DNA levels for hepatocellular carcinoma.
1. LLOVET JM, BRU C, BRUIX J. Prognosis of hepatocellular carcinoma: International Journal of Infectious Diseases. 2018; 67: 92-7.
The bclc staging classification. Seminars in liver disease. 1999; 19. 26. Xu H, Zhu X, Xu Z, Hu Y, Bo S, Xing T, et al. Non-invasive analysis of
2. Lang H, Sotiropoulos GC, Brokalaki EI, Schmitz KJ, Bertona C, Meyer G, genomic copy number variation in patients with hepatocellular
et al. Survival and recurrence rates after resection for hepatocellular carcinoma by next generation DNA sequencing. Journal of Cancer. 2015;
carcinoma in noncirrhotic livers. Journal of the american college of 6: 247-53.
surgeons. 2007; 205: 27-36.

http://www.ijbs.com
Int. J. Biol. Sci. 2020, Vol. 16 1561
27. Cai Z, Chen G, Zeng Y, Dong X, Li Z, Huang Y, et al. Comprehensive 47. Holmila R, Sklias A, Muller DC, Esposti DD, Guilloreau P, McKay J, et al.
liquid profiling of circulating tumor DNA and protein biomarkers in Targeted deep sequencing of plasma circulating cell-free DNA reveals
long-term follow-up patients with hepatocellular carcinoma. Clin Cancer Vimentin and Fibulin 1 as potential epigenetic biomarkers for
Res. 2019; 25: 5284-94. hepatocellular carcinoma. PLoS One. 2017; 12: e0174265.
28. Oh CR, Kong SY, Im HS, Kim HJ, Kim MK, Yoon KA, et al. 48. Zhang PJ, Wen XY, Gu F, Deng XX, Li J, Dong J, et al. Methylation
Genome-wide copy number alteration and VEGFA amplification of profiling of serum DNA from hepatocellular carcinoma patients using an
circulating cell-free DNA as a biomarker in advanced hepatocellular infinium human methylation 450 beadchip. Hepatology International.
carcinoma patients treated with Sorafenib. BMC Cancer. 2019; 19: 292. 2013; 7: 893-900.
29. Jiang P, Chan CWM, Chan KCA, Cheng SH, Wong J, Wong VW-S, et al. 49. Han LY, Fan YC, Mu NN, Gao S, Li F, Ji XF, et al. Aberrant DNA
Lengthening and shortening of plasma DNA in hepatocellular methylation of g-protein-coupled bile acid receptor gpbar1 (TGR5) is a
carcinoma patients. PROCEEDINGS OF THE NATIONAL ACADEMY potential biomarker for hepatitis b virus associated hepatocellular
OF SCIENCES OF THE UNITED STATES OF AMERICA. 2015(E): 1317– carcinoma. International Journal of Medical Sciences. 2014; 11: 164-71.
25. 50. Ji XF, Fan YC, Gao S, Yang Y, Zhang JJ, Wang K. MT1M and MT1G
30. Huang A, Zhang X, Zhou SL, Cao Y, Huang XW, Fan J, et al. Plasma promoter methylation as biomarkers for hepatocellular carcinoma.
Circulating Cell-free DNA Integrity as a Promising Biomarker for World J Gastroenterology. 2014; 20: 4723-9.
Diagnosis and Surveillance in Patients with Hepatocellular Carcinoma. 51. Xu RH, Wei W, Krawczyk M, Wang W, Luo H, Flagg K, et al. Circulating
Journal of Cancer. 2016; 7: 1798-803. tumour DNA methylation markers for diagnosis and prognosis for
31. Jiang P, Sun K, Tong YK, Cheng SH, Cheng THT, Heung MMS, et al. diagnosisandprognosis of hepatocellular carcinoma. Nature Materials.
Preferred end coordinates and somatic variants as signatures of 2017; 16.
circulating tumor DNA associated with hepatocellular carcinoma. 52. Cheng J, Holland-Letz T, Wallwiener M, Surowy H, Cuk K, Schott S, et
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF al. Circulating free DNA integrity and concentration as independent
THE UNITED STATES OF AMERICA. 2018; 115. prognostic markers in metastatic breast cancer. Breast Cancer Res Treat.
32. Lim HY, Merle P, Weiss KH, Yau T, Ross P, Mazzaferro V, et al. Phase II 2018; 169: 69-82.
Studies with Refametinib or Refametinib plus Sorafenib in Patients with 53. Cohen JD, Javed AA, Thoburn C, Wong F, Tie J, Gibbs P, et al. Combined
RAS-Mutated Hepatocellular Carcinoma. Clinical Cancer Research. 2018; circulating tumor DNA and protein biomarker-based liquid biopsy for
24: 19. the earlier detection of pancreatic cancers. Proc Natl Acad Sci U S A.
33. Piciocchi M, Cardin R, Vitale A, Vanin V, Giacomin A, Pozzan C, et al. 2017; 114: 10202-7.
Circulating free DNA in the progression of liver damage to 54. Lee SE, Chang SH, Kim WY, Hwang TS, Kim WY, Han HS. Frequent
hepatocellular carcinoma. Hepatology International. 2013; 7: 1050-7. somatic TERT promoter mutations and CTNNB1 mutations in
34. Ng CKY, Di Costanzo GG, Tosti N, Paradiso V, Coto-Llerena M, hepatocellular carcinoma. Oncotarget. 2016; 7.
Roscigno G, et al. Genetic profiling using plasma-derived cell-free DNA 55. Network. CGAR. Comprehensive and Integrative Genomic
in therapy-naive hepatocellular carcinoma patients: a pilot study. Annals Characterization of Hepatocellular Carcinoma. Cell. 2017; 169(e23):
of Oncology 2018; 29: 1286-91. 1327-41.
35. Huang A, Zhao X, Yang XR, Li FQ, Zhou XL, Wu K, et al. Circumventing 56. Whittaker S, Marais R, Zhu AX. The role of signaling pathways in the
intratumoral heterogeneity to identify potential therapeutic targets in development and treatment of hepatocellular carcinoma. Oncogene.
hepatocellular carcinoma. Journal of Hepatology. 2017; 67: 293-301. 2010; 29: 4989-5005.
36. Howell J, Atkinson SR, Pinato DJ, Knapp S, Ward C, Minisini R, et al. 57. Gollob JA, Wilhelm S, Carter C, Kelley SL. Role of Raf kinase in cancer:
Identification of mutations in circulating cell-free tumour DNA as a therapeutic potential of targeting the Raf/MEK/ERK signal transduction
biomarker in hepatocellular carcinoma. European Journal of Cancer. pathway. Semin Oncol. 2006; 33: 392-406.
2019; 116: 56-66. 58. Downward J. Targeting RAS signalling pathways in cancer therapy. Nat
37. He G, Chen Y, Zhu C, Zhou J, Xie X, Fei R, et al. Application of plasma Rev Cancer. 2003; 3: 11-22.
circulating cell free DNA detection to the molecular diagnosis of 59. Pinyol R, Tovar V, Llovet JM. TERT promoter mutations: gatekeeper and
hepatocellular carcinoma. American Journal of Translational Research. driver of hepatocellular carcinoma. Journal of Hepatology. 2014; 61:
2019. 685-7.
38. Xiong Y, Xie CR, Zhang S, Chen J, Yin ZY. Detection of a novel panel of 60. Hussain SP, Schwank J, Staib F, Wang XW, Harris CC. TP53 mutations
somatic mutations in plasma cell-free DNA and its diagnostic value in and hepatocellular carcinoma: insights into the etiology and
hepatocellular carcinoma. Cancer Management and Research. 2019; 11: pathogenesis of liver cancer. Oncogene. 2007; 26: 2166-76.
5745-56. 61. Zhan T, Rindtorff N, Boutros M. Wnt signaling in cancer. Oncogene.
39. Song CX, Yin SL, Ma L, Wheeler A, Chen Y, Zhang Y, et al. 2017; 36: 1461-73.
5-hydroxymethylcytosine signatures in cell-free DNA provide 62. Kimelman D, Xu W. β-Catenin destruction complex: insights and
information about tumor types and stages. Cell Research. 2017; 27: questions from a structural perspective. Oncogene. 2006; 25: 7482-91.
1231-42. 63. Abitbol S, Dahmani R, Coulouarn C, Ragazzon B, Mlecnik B, Senni N, et
40. Li W, Zhang X, Lu X, You L, Song Y, Luo Z, et al. al. AXIN deficiency in human and mouse hepatocytes induces
5-Hydroxymethylcytosine signatures in circulating cell-free DNA as hepatocellular carcinoma in the absence of beta-catenin activation.
diagnostic biomarkers for human cancers. Cell Research. 2017; 27: Journal of Hepatology. 2018; 68: 1203-13.
1243-57. 64. Sun X, Wang SC, Wei Y, Luo X, Jia Y, Li L, et al. Arid1a Has
41. Cai J, Chen L, Zhang Z, Zhang X, Lu X, Liu W, et al. Genome-wide Context-Dependent Oncogenic and Tumor Suppressor Functions in
mapping of 5-hydroxymethylcytosines in circulating cell-free DNA as a Liver Cancer. Cancer Cell. 2017; 32(e6): 574-89.
non-invasive approach for early detection of hepatocellular carcinoma. 65. Liu J, Jiang J, Mo J, Liu D, Cao D, Wang H, et al. Global DNA
Gut. 2019; 68: 2195-205. 5-Hydroxymethylcytosine and 5-Formylcytosine Contents Are
42. Li W, Li Q, Kang S, Same M, Zhou Y, Sun C, et al. Cancerdetector: Decreased in the Early Stage of Hepatocellular Carcinoma. Hepatology.
Ultrasensitive and non-invasive cancer detection at the resolution of 2019; 69: 196-208.
individual reads using cell-free DNA methylation sequencing data. 66. Wysokinski D, Blasiak J, Pawlowska E. Role of RUNX2 in Breast
Nucleic Acids Research. 2018; 46: e89. Carcinogenesis. International journal of molecular sciences. 2015; 16:
43. Wen L, Li J, Guo H, Liu X, Zheng S, Zhang D, et al. Genome-scale 20969-93.
detection of hypermethylated CpG islands in circulating cell-free DNA 67. Bradley JE, Chan JM, Hagood JS. Effect of the GPI anchor of human
of hepatocellular carcinoma patients. cell Research. 2015; 15: 1-15. Thy-1 on antibody recognition and function. Lab Invest. 2013; 93: 365-74.
44. Oussalah A, Rischer S, Bensenane M, Conroy G, Filhine-Tresarrieu P, 68. Angata K, Fukuda M. Polysialyltransferases: major players in polysialic
Debard R, et al. Plasma mSEPT9: A novel circulating cell-free acid synthesis on the neural cell adhesion molecule. Biochimie. 2003; 85:
DNA-based epigenetic biomarker to diagnose hepatocellular carcinoma. 195-206.
EBioMedicine. 2018: 138-47. 69. Si M, Lang J. The roles of metallothioneins in carcinogenesis. Journal of
45. Huang G, Krocker JD, Kirk JL, Merwat SN, Ju H, Soloway RD, et al. Hematology & Oncology. 2018; 11.
Evaluation of INK4A promoter methylation using pyrosequencing and 70. Hunt TW, Fields TA, Casey PJ, Peralta EG. RGS10 is a selective activator
circulating cell-free DNA from patients with hepatocellular carcinoma. of Gαi GTPase activity. Nature. 1996; 383.
Clinical Chemistry Labratory Medicine. 2014; 52: 899-909. 71. Byun Y, Chen F, Chang R, Trivedi M, Green KJ, Cryns VL. Caspase
46. Vaca-Paniagua F, Oliver J, Nogueira da Costa A, Merle P, McKay J, cleavage of vimentin disrupts intermediate filaments and promotes
Herceg Z, et al. Targeted deep DNA methylation analysis of circulating apoptosis. Cell Death and Differentiation. 2001; 8.
cell-free DNA in plasma using massively parallel semiconductor
sequencing. Epigenomics. 2015; 7: 353-62.

http://www.ijbs.com
Int. J. Biol. Sci. 2020, Vol. 16 1562
72. Pols TWH, Noriega LG, Nomura M, Auwerx J, Schoonjans K. The bile
acid membrane receptor TGR5: A valuable metabolic target. Digestive
Diseases. 2011; 29: 37-44.
73. Estey MP, Di Ciano-Oliveira C, Froese CD, Bejide MT, Trimble WS.
Distinct roles of septins in cytokinesis: SEPT9 mediates midbody
abscission. The Journal of Cell Biology. 2010; 191: 741-9.
74. Roussel MF. The INK4 family of cell cycle inhibitors in cancer.
Oncogene. 1999; 18.
75. Piscaglia F, Dudás J, Knittel T, Di Rocco P, Kobold D, Saile B, et al.
Expression of ECM proteins fibulin-1 and -2 in acute and chronic liver
disease and in cultured rat liver cells. Cell and Tissue Research. 2009;
337: 449-62.
76. Cohen JD, Li L, Wang YX, Thoburn C, Afsari B, Danilova L, et al.
Detection and localization of surgically resectable cancers with a
multi-analyte blood test. Science. 2018: 926–30.
77. Xu R-h, Wei W, Krawczyk M, Wang W, Luo H, Flagg K, et al. Circulating
tumour DNA methylation markers for diagnosis and prognosis of
hepatocellular carcinoma. Nature Materials. 2017; 16: 1155.
78. Phallen J, Sausen M, Adleff V, Leal A, Hruban C, White J, et al. Direct
detection of early-stage cancers using circulating tumor DNA. Science
translational medicine. 2017: eaan2415
79. Heitzer E, Haque IS, Roberts CES, Speicher MR. Current and future
perspectives of liquid biopsies in genomics-driven oncology. Nature
Reviews Genetics. 2019; 20: 71-88.
80. Raoul JL, Forner A, Bolondi L, Cheung TT, Kloeckner R, de Baere T.
Updated use of TACE for hepatocellular carcinoma treatment: How and
when to use it based on clinical evidence. Cancer Treatment Reviews.
2019; 72: 28-36.
81. Llovet JM, Ricci S, Mazzaferro V, Hilgard P, Gane E, Blanc J-F, et al.
Sorafenib in advanced hepatocellular carcinoma. The New England
Journal of Medicine. 2008.
82. Scherer F, Kurtz DM, Newman AM, Stehr H, Craig AF, Esfahani MS, et
al. Distinct biological subtypes and patterns of genome evolution in
lymphoma revealed by circulating tumor DNA. Science Translational
Medicine. 2016; 8: 364ra155.
83. Lyu M, Zhou J, Ning K, Ying B. The diagnostic value of circulating tumor
cells and ctDNA for gene mutations in lung cancer. OncoTargets and
Therapy. 2019; 12: 2539-52.
84. Bi WL, Hosny A, Schabath MB, Giger ML, Birkbak NJ, Mehrtash A, et al.
Artificial intelligence in cancer imaging: Clinical challenges and
applications. CA Cancer J Clin. 2019; 69: 127-57.
85. Yin CQ, Yuan CH, Qu Z, Guan Q, Chen H, Wang FB. Liquid biopsy of
hepatocellular carcinoma: Circulating tumor-derived biomarkers.
Disease Markers. 2016; 2016: 1427849.

http://www.ijbs.com