cancers

Systematic Review
Candidate Genes and Pathways in Cervical Cancer:
A Systematic Review and Integrated Bioinformatic Analysis
Marjanu Hikmah Elias 1 , Srijit Das 2 and Nazefah Abdul Hamid 1, *

1 Department of Basic Medical Sciences I, Faculty of Medicine & Health Sciences,

Universiti Sains Islam Malaysia, Nilai 71800, Malaysia
2 Department of Human & Clinical Anatomy, College of Medicine and Health Sciences,
Sultan Qaboos University, Muscat 123, Oman
* Correspondence: [email protected]

Simple Summary: Cervical cancer is the fourth most common cancer among women worldwide.
Although many recommendations on the screening, diagnosis, and treatment of cervical cancer have
been established, no comprehensive molecular mechanism for cervical cancer has been determined.
Hence, this systematic review and integrated bioinformatic analysis provides new insight into the
key genes and pathways involved in the pathogenesis of cervical cancer and, thus, can be beneficial
for developing better screening and treatment strategies for cervical cancer.

Abstract: Cervical cancer is the leading cause of cancer-related death among women in developing
countries. However, no comprehensive molecular mechanism for cervical cancer has been established,
as many studies were small-cohort studies conducted with small sample sizes. A thorough literature
search was performed using the PubMed, Scopus, EBSCOhost, and Science Direct databases. Medical
Subject Heading (MeSH) terms such as “Uterine Cervical Neoplasms” and “gene expression” were
used as the keywords in all fields. A total of 4027 studies were retrieved, and only clinical studies,
which used the microarray method to identify differentially expressed genes (DEGs) in the cervical
tissue of cervical cancer patients, were selected. Following the screening, 6 studies were selected
and 1128 DEGs were extracted from the data. Sixty-two differentially expressed genes from at least
two studies were selected for further analysis by DAVID, STRING, and Cytoscape software. In
Citation: Elias, M.H.; Das, S.; Abdul
cervical cancer pathogenesis, three significant clusters with high intermolecular interactions from
Hamid, N. Candidate Genes and
the Protein–Protein Interaction (PPI) network complex revealed three major molecular mechanisms,
Pathways in Cervical Cancer: A
including cell signaling, cell cycle, and cell differentiation. Subsequently, eight genes were chosen
Systematic Review and Integrated
Bioinformatic Analysis. Cancers 2023,
as the candidate genes based on their involvement in the relevant gene ontology (GO) and their
15, 853. https://doi.org/10.3390/ interaction with other genes in the PPI network through undirected first neighbor nodes. The present
cancers15030853 systematic review improves our understanding of the molecular mechanism of cervical cancer and
the proposed genes that can be used to expand the biomarker panel in the screening for cervical
Academic Editors: Giorgio Treglia
cancer. The targeted genes may be beneficial for the development of better treatment strategies.
and Ursula Maria Vogl

Received: 19 December 2022 Keywords: cervical cancer; differentially expressed gene; molecular pathway; gene ontology
Revised: 20 January 2023
Accepted: 23 January 2023
Published: 30 January 2023

1. Introduction
Cervical cancer is the fourth most common cancer in women worldwide, with an
Copyright: © 2023 by the authors.
estimated incidence of 570,000 and 311,000 related deaths in 2018 [1]. Although the inci-
Licensee MDPI, Basel, Switzerland. dence of cervical cancer has declined in the United States, it remains the leading cause of
This article is an open access article cancer-related death in developing countries [2]. Thus, many recommendations on the
distributed under the terms and screening, diagnosis, and treatment of cervical cancer have been established [2–4]. A recent
conditions of the Creative Commons recommendation on multiple-agent chemotherapy regimens that include targeted thera-
Attribution (CC BY) license (https:// pies and immunotherapy regimens in combination with the existing first- and second-line
creativecommons.org/licenses/by/ treatment options has improved the outcomes in some cervical cancer patients [3]. How-
4.0/). ever, cervical cancer is a multifactorial disease that involves complex pathophysiology and

Cancers 2023, 15, 853. https://doi.org/10.3390/cancers15030853 https://www.mdpi.com/journal/cancers

Cancers 2023, 15, 853 2 of 13

molecular pathogenesis. Thus, further studies are still crucial to providing comprehensive
information on the mechanism of cervical cancer pathogenesis that is critical in developing
better screening, diagnosis, and treatment of cervical cancer.
Several molecular mechanisms have been proposed for cervical cancer involving
genomic variations [5], regulation of mRNA expression [6], and epigenetic changes [7].
Various pathways have been identified to be involved in cervical cancer through the regu-
lation of gene expression including MAPK, mTOR, PI3K-Akt, the Ras signaling pathway,
immune response, inflammation, DNA synthesis, cell proliferation, and many others [8,9].
Identifying the DEGs and their key pathways and functions can improve our understand-
ing of the molecular mechanisms that occur in cervical cancer pathogenesis. Therefore,
many studies on gene-expression profiling have been conducted on cervical cancer patients,
and hundreds of DEGs have been identified. From these studies, many potential biomark-
ers have been proposed, such as FBL1, ANT3 [10], RBBP6 [11], TMEM45A, SERPINB5,
p16INK4A [12], and many more. However, these studies only focused on a single cohort
from different populations, with a small sample size, and employed different methods to
identify the DEGs. Thus, these individual results could not robustly represent the pathogen-
esis of cervical cancer due to various biases. Hence, this systematic review was performed
to identify the pattern of gene expression, functions, interactions, and critical pathways
with minimal bias in the cervical tissue of cervical cancer patients across populations via
integrated bioinformatic analysis.

2. Materials and Methods

2.1. Search Strategy
The systematic review was conducted according to the PRISMA guidelines. This
systematic review was registered with OpenScience Framework. A comprehensive liter-
ature search was carried out using the PubMed, Scopus, EBSCOhost, and Science Direct
databases. Related research papers published up to 15 February 2022 were identified.
Medical Subject Heading (MeSH) terms such as “uterine cervical neoplasms” and “gene
expression” were used as the keywords in all fields. Synonyms for the keywords were
generated through MeSH terms from the Cochrane Library. Additional text terms were
found by assessing the collected review articles. The search strategy involved a combination
(“AND”) of the following sets of keywords: (1) “uterine cervical neoplasms” OR “cervi-
cal intraepithelial neoplasia” OR “cervical cancer” and (2) “gene expression.” Additional
references were identified from the bibliographies of the retrieved studies.

2.2. Inclusion Criteria

Case–control, cross-sectional, and prospective observational studies with abstracts
investigating the DEGs in the cervical tissue of cervical cancer patients were included.
Only clinical studies using the microarray method to identify differentially expressed
genes (DEGs) in the cervical tissue of cervical cancer patients were included to ensure
the homogeneity of the data. Studies reporting a list of DEGs were included in this
systematic review.

2.3. Exclusion Criteria

Publications without primary data, such as editorials, case reports, conference proceed-
ings, and narrative review articles, were excluded. In silico, in vitro, and in vivo studies
were excluded. Intervention studies for cervical cancer treatment and those using blood
and non-cervical tissue samples were excluded. Studies that utilized other gene-expression
methods, such as quantitative PCR, were not selected to avoid bias in gene selection. Next-
Generation Sequencing (NGS) studies were excluded to avoid allele expression biases.
Studies that did not include the list of DEGs in their report, supplementary materials, or
links to any other sources were also excluded.
 selection. Next-Generation Sequencing (NGS) studies were excluded to avoid all
pression biases. Studies that did not include the list of DEGs in their report, suppl
tary materials, or links to any other sources were also excluded.

Cancers 2023, 15, 853 2.4. Screening of Articles for Eligibility 3 of 13

Three phases of screening were performed on articles recovered from all reso
Duplicates were removed, and all articles with non-relevant titles were excluded
2.4. Screening of Articles for Eligibility
first phase. The abstracts of the remaining articles were examined, and articles th
Three
notphases of screening
meet the inclusionwere performed
criteria on articles
were excluded recovered
from fromphase.
the second all resources.
Finally, the fu
Duplicates were removed, and all articles with non-relevant titles were excluded in the
of the remaining articles were reviewed thoroughly. Systematic reviews; meta-an
first phase. The abstracts of the remaining articles were examined, and articles that did
in vitro, in vivo, and in silico articles; and articles that did not meet the inclusion c
not meet the inclusion criteria were excluded from the second phase. Finally, the full texts
were
of the remaining excluded
articles in thisreviewed
were third phase. All theSystematic
thoroughly. authors were involved
reviews; in the screening,
meta-analyses;
in vitro, tion, and
in vivo, data
and extraction
in silico phases.
articles; Figurethat
and articles 1 shows
did notthe PRISMA
meet flow diagram
the inclusion criteria summa
the article-sorting
were excluded process,
in this third phase. andauthors
All the the reasons for article
were involved in elimination.
the screening, selection,
and data extraction phases. Figure 1 shows the PRISMA flow diagram summarizing the
article-sorting process, and the reasons for article elimination.

Figure 1. PRISMA flow diagram for studies’ selection in this systematic review.
Figure 1. PRISMA flow diagram for studies’ selection in this systematic review.
2.5. Data Extraction
Data were
2.5. extracted
Data from the studies that fulfilled the inclusion criteria. All the authors
Extraction
participated in extracting the data. A data collection form was used to standardize the data
Data were extracted from the studies that fulfilled the inclusion criteria. All t
collection, and all data extraction was performed independently. Any disagreements were
thors participated
discussed, decisions in extracting
were made themajority
based on the data. A consensus,
data collection form was
and records used to stand
of reasons
for rejection were kept. The collected data are as follows: (1) author’s name, (2) article title,
(3) study design, (4) sample size, (5) type of sample, (6) gene-expression method, (7) list of
DEGs, and (8) conclusion.
Cancers 2023, 15, 853 4 of 13

2.6. Study Quality

The study quality of each paper was evaluated independently by NAH and MHE
using the Joanna Briggs Institute critical appraisal tools [13]. The results of the quality of
the study were validated by MHE and NAH. An overall score of less than 50% was used to
rate a paper as a low-quality (high risk of bias) paper. If the overall score was 50–69%, it
was rated as a moderate-quality (moderate risk of bias) paper, and if the overall score was
more than 69%, it was rated as a high-quality (low risk of bias) paper.

2.7. Venn Diagram Analysis

All the DEGs were extracted and listed according to the studies. Then, a Venn diagram
analysis was carried out using Bioinformatics & Evolutionary Genomics software [14]. The
Venn diagram analysis was performed to identify the common DEGs between studies.
Differentially expressed genes in at least two studies were selected for further analysis.

2.8. Protein–Protein Interaction (PPI) Network, Clustering, and Visualization

All selected DEGs from the Venn diagram analysis were pooled and analyzed through
a PPI functional-enrichment analysis via STRING software (version 11.5, STRING Consor-
tium 2022, Zurich, Switzerland, https://string-db.org/ accessed on 28 October 2022) to
identify the PPI network [15]. The results from STRING were exported into Cytoscape soft-
ware (version 3.8.0, Cytoscape Consortium, San Diego, CA, USA, http://www.cytoscape.
org/, accessed on 4 November 2022) to visualize the molecular interaction networks and to
integrate the gene-expression profiles of the DEGs [16]. A module analysis of the target
network and protein clustering were performed using the Cytoscape MCODE plug-in
(degree cut-off = 2, node score cut-off = 0.2, node density cut-off = 0.1, K-score = 2, and max
depth = 100). The significantly enriched gene ontology was identified by analyzing the list
of genes in each cluster using DAVID software (version 2021, DAVID Bioinformatic team,
Frederick, MD, USA).

2.9. Gene Ontology (GO) and Pathway Enrichment Analysis

All the genes in each cluster were analyzed using the Database for Annotation, Visual-
ization, and Integrated Discovery (DAVID) to discover the gene ontology that exhibited
significant functional annotation enrichment related to cervical cancer pathogenesis [17].
Finally, the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway was utilized to
expose the involvement of genes in the pathway related to cervical cancer [18].

3. Results
The four databases generated 4027 potentially relevant studies from the keywords.
Based on the title, 307 duplicates were removed, and the abstracts of the other 3720 were
screened. Upon screening the abstracts, 3681 papers were excluded. The full text of the
other 24 papers was then retrieved. After a thorough review of the full text, 18 articles that
did not meet the inclusion and exclusion criteria were excluded. Finally, six studies were
selected for this systematic review. These six studies were published between the years
2006 and 2021. Prevention of sampling bias was assured through homogenized sampling
by adhering strictly to the specific inclusion and exclusion criteria. Only studies that
performed microarray analyses were chosen, and the sample size was between 2 and 25.
Table 1 shows a summary of the characteristics of the included studies.
Cancers 2023, 15, 853 5 of 13

Table 1. Summary of the characteristics of the included studies.

Gene-
Author, Year Study Sample Size Upregulated Downregulated
Country Sampling Expression
(References) Design (n) Genes Genes
Analysis
Annapurna
Cervical 7 patients,
et al., 2021 India Case–control Microarray 78 26
tissue 3 normal
[19]
Kim et al., Cervical 28 patients,
Korea Case–control Microarray 208 201
2013 [20] tissue 17 control
Rajkumar
Cervical 28 patients, 5
et al., 2011 India Case–control Microarray 47 19
tissue normal
[21]
Miyatake
Cross- Cervical
et al., 2007 Japan 2 patients Microarray 8 14
sectional tissue
[22]
Gius et al., Cross- Cervical
USA 85 patients Microarray 536
2007 [23] sectional tissue
Wong et al., Cervical 29 patients,
Hong Kong Case–control Microarray 19 83
2006 [24] tissue 18 controls

3.1. Patient Recruitment and Sample Collection Details

Sample collection was described in all studies. Kim et al. (2013), Miyatake et al.
(2007), and Wong et al. (2006) stated that the staging of cervical cancer was carried out
according to the International Federation of Gynaecology and Obstetrics (FIGO) crite-
ria [20,22,24]. However, not all studies elaborated on the demographic profile of their
patients. While the other five studies collected biopsies, Rajkumar et al. (2011) collected
the archive-extracted total RNA from the biobank [21]. Normal cervical tissue biopsies
were collected from women who underwent hysterectomy for other gynecological-related
problems [19,21,24], from uterine leiomyoma patients [20], and from normal tissue adjacent
to cervical squamous epithelia [22,23].

3.2. Study Quality

A detailed quality assessment of the included studies is listed in Supplementary Table S1.
The included studies comprised four high-quality (low risk of bias) studies and two
moderate-quality (moderate risk of bias) studies.

3.3. Identification of DEGs in Cervical Cancer

After removing duplicates, 1128 DEGs were extracted from all the selected studies. In
the studies by Annapurna et al. (2021), Kim et al. (2013), Rajkumar et al. (2011), Miyatake
et al. (2007), and Wong et al. (2006), only genes with an expression level of more than two
or less than a negative two-fold change are identified as DEGs [19–22,24]. However, in
the study by Gius et al. (2007), instead of using the fold change as a parameter to select
DEGs, this study uses the calculated false discovery rate [23]. A list of 104, 409, 66, 22, 536,
and 102 DEGs were reported in Annapurna et al. (2021), Kim et al. (2013), Rajkumar et al.
(2011), Miyatake et al. (2007), Gius et al. (2007), and Wong et al. (2006), respectively [19–24].

3.4. Common DEGs among the Studies Identified via Venn Diagram Analysis
The common DEGs between studies were identified, in which CDH3 and CDKN2A
were differentially expressed in four studies (Annapurna et al., 2021; Gius et al., 2007;
Rajkumar et al., 2011; Wong et al., 2006) [19,21,23,24]. Four other genes (BST2, PLSCR1,
SPINK5, PLOD2) were common in three studies, and sixty-two genes were commonly
expressed in two studies. Figure 2 shows the Venn diagram result for the DEGs from
Cancers 2023, 15, 853 6 of 13

Cancers 2023,
Cancers 15, x15, x
2023, 6 of 613of 13
all the included studies. All 68 genes differentially expressed in at least two studies
(Supplementary Table S2) were selected for further analysis.

Figure 2. Venn diagram of the DEGs from all the selected studies.
Figure 2. Venn diagram
Figure diagramof ofthe
theDEGs
DEGsfrom
fromallall
the selected
the studies.
selected studies.
3.5.3.5.
Identification of Key Candidate GenesGenes
and Pathways via Protein–Protein Interaction (PPI) (PPI)
3.5. Identification
Identification ofofKeyKey Candidate
Candidate Genes and and Pathways
Pathways viavia Protein–Protein
Protein–Protein Interaction
Interaction (PPI)
Network
Network and Modular Analysis
Network and ModularAnalysis
and Modular Analysis
Sixty-eight
Sixty-eight proteins fromfrom
proteins the selected DEGsDEGs
theselected
selected were filtered
were into a PPI
filtered network complex complex
Sixty-eight proteins from the DEGs were filtered intointo
a PPIa PPI network
network complex
containing
containing 68 nodes
68 nodesand
nodes and 181
and 181edges
181edges with
edgeswith a PPI
withaaPPI enrichment
PPIenrichment p-value
enrichmentp-value of
p-value <1 × 10 -16 and − an16 and an
×-1610and
containing 68 of of
<1<1
× 10 an
average
average local clustering coefficient of 0.602. The network’s data were transferred from
average local clusteringcoefficient
local clustering coefficientofof0.602.
0.602.The The network’s
network’s datadata
were were transferred
transferred fromfrom
STRING to Cytoscape software to visualize the molecular interaction networks. By utiliz-
STRING
STRING to to Cytoscape software
software totovisualize
visualizethe themolecular
molecularinteraction
interactionnetworks.
networks. ByBy utilizing
utiliz-
ing the Molecular Complex Detection Algorithm (MCODE), three significant modules
the
ing Molecular
the Molecular Complex
Complex Detection
Detection Algorithm
Algorithm (MCODE),
(MCODE), three
threesignificant
significantmodules
modulesfrom
from the PPI network complex were discovered. Figure 3 shows the three significant clus-
the
from PPI
thenetwork complex
PPI network complexwere discovered.
were discovered. Figure
Figure33showsshows the threesignificant
the three significantclus-clusters
ters created from the PPI network complex generated from the DEGs in cervical cancer
ters created
created fromfrom
the PPIthenetwork
PPI network
complexcomplex generated
generated fromfrom the DEGs
the DEGs in cervical
in cervical cancer
cancer patients.
patients. The functional annotation clustering shows that cluster 1 comprises 20 nodes and
patients. The functional annotation clustering shows that cluster 1 comprises 20 nodes 58 and
58 The
edgesfunctional annotation
(score = 6.105), clustering
while cluster showsofthat
2 consists cluster
7 nodes and113 comprises
edges (score20 nodes
= 4.33),and
and edges
58 edges
(score (score = 6.105),
= 6.105),ofwhile while
cluster cluster 2 consists of 7 nodes and 13 edges (score = 4.33), and
cluster 3 consists 4 nodes and 62edges
consists of =7 4).
(score nodes and 13 edges (score = 4.33), and cluster
cluster 3 consists of 4 nodes and 6
3 consists of 4 nodes and 6 edges (score = 4).edges (score = 4).

Figure 3. PPI network of the DEGs collected from the selected studies.
Cancers 2023, 15, 853 7 of 13

3.6. Gene Ontology (GO) and Pathway Enrichment Analysis of the Identified Clusters
The GO and pathway enrichment analyses showed that the DEGs in cluster 1 were
primarily located in the nucleoplasm and cytosol. These DEGs in cluster 1 involve the cell
cycle and regulation of transcription. Their molecular functions include protein binding,
DNA binding, ATP binding, and transcription factor binding. The DEGs in cluster 2 are
primarily located in the cytoplasm and regulate cell adhesion, migration, and inflammatory
and immune responses. The molecular function of DEGs in cluster 2 includes extracellular
matrix binding, cytokine activity, and fibronectin binding.
The DEGs in cluster 3 were mainly situated in the cytosol and cornified envelope.
These DEGs are involved in protein heterotetramerization, peptide cross-linking, and ep-
ithelial cell differentiation. The molecular function of DEGs in cluster 3 includes structural
constituents of the epidermis. Table 2 summarizes the functional annotation clustering of
the DEGs.

Table 2. Functional annotation clustering of the DEGs with highlighted candidate genes.

Cluster Term Description Genes p-Value

EGR1, CDKN2A, FOS, KLF4, SMC4, ESR1, CKS1B,
1 CC_ GO:0005654 Nucleoplasm 6.06 ×10−6
RAD51C, IFI16, MCM4, MCM6, IRF9, MCM2
BST2, CCNB2, RAD51C, IFI16, CDKN2A, FOS,
CC_GO:0005829 Cytosol 1.96 × 10−2
KLF4, SMC4, ESR1, IRF9
CC_GO:0000785 Chromatin EGR1, FOS, KLF4, ESR1, IRF9, MCM2 1.32 × 10−3
Positive regulation of
BP_GO:0045944 transcription from RNA EGR1, IFI16, CDKN2A, FOS, KLF4, ESR1, IRF9 3.96 × 10−4
polymerase II promoter
BP_GO:0007049 Cell cycle CDKN2A, MCM4, MCM6, SMC4, MCM2, CKS1B 1.11 × 10−5
Positive regulation of
BP_GO:0045893 EGR1, CDKN2A, SPP1, FOS, KLF4, ESR1 3.13 × 10−4
transcription, DNA-templated
MF_ GO:0005524 ATP binding RAD51C, MCM4, MCM6, SMC4, MCM2 4.50 × 10−2
MF_ GO:0003697 Single-stranded DNA binding MCM4, MCM6, SMC4, MCM2 1.41 × 10−4
MF_GO:0008134 Transcription factor binding IFI16, CDKN2A, FOS, ESR1 8.12 × 10−4
hsa04110 Cell cycle CCNB2, CDKN2A, MCM4, MCM6, MCM2 8.53 × 10−5
2 CC_GO:0005737 Cytoplasm IL1RN, CXCL12, IL18, PTGS2, VEGFA 0.049
Positive regulation of cell
BP_ GO:0045785 CXCL12, ITGAV, VEGFA 1.37 × 10−4
adhesion
Positive regulation of cell
BP_GO:0030335 CXCL12, ITGAV, VEGFA 0.002
migration
BP_ GO:0006954 Inflammatory response IL1RN, IL18, PTGS2 0.006
MF_ GO:0005125 Cytokine activity IL1RN, IL18, VEGFA 0.002
MF_ GO:0050840 Extracellular matrix binding ITGAV, VEGFA 0.009
hsa05165 Human papillomavirus infection ITGAV, PTGS2, VEGFA 0.022
3 CC_GO:0005829 Cytosol KRT1, SPINK5, DSG1, KRT10 0.019
CC_ GO:0045095 Keratin filament KRT1, KRT10 0.015
BP_ GO:0051290 Protein heterotetramerization KRT1, KRT10 0.002
BP_ GO:0018149 Peptide cross-linking KRT1, KRT10 0.005
Structural constituent of
MF_ GO:0030280 KRT1, KRT10 0.006
epidermis
 Cancers 2023, 15, x 8 of 13

Structural constituent of epi-

Cancers 2023,MF_ GO:0030280
15, 853 KRT1, KRT10 0.006
8 of 13
dermis

3.7. Selection of Candidate Genes in Cervical Cancer Pathogenesis

3.7. Selection
From theof Candidate
PPI networkGenes in Cervical
and Cancer annotation
the functional Pathogenesisclustering, eight genes were
From
chosen as the candidate
PPI network andbased
genes the functional annotation in
on their involvement clustering,
the relevanteight
GOgenes were
and their
chosen as the
interaction candidate
with genes
other genes in based
the PPIonnetwork
their involvement in the relevant
through undirected GO and
first neighbor their
nodes
interaction withcandidate
(Figure 4). The other genes in include
genes the PPI CDNK2A,
network through
VEGFA,undirected
PTGS2, MCM2, first MCM4,
neighbor nodes
MCM6,
(Figure 4). The
KRT1, and candidate
KRT10. Fromgenes CDNK2A, VEGFA, PTGS2,
include CDNK2A,
the analysis, and PTGS2MCM2, MCM4,closely.
interacted MCM6,
KRT1, and KRT10. From the analysis, CDNK2A, VEGFA, and PTGS2 interacted
Meanwhile, the MCM and KRT families interact less with other clusters. Further details closely.
Meanwhile, MCM KRT
were extracted for all the candidate genes. Their expression levels in cervical cancerdetails
the and families interact less with other clusters. Further sam-
were
ples extracted for allannotation
and functional the candidate genes.
of each Their
gene setexpression
are shown levels in cervical
in Table 3. cancer samples
and functional annotation of each gene set are shown in Table 3.

Figure 4. Protein interaction through undirected first neighbor of the candidate genes (highlighted in
yellow) from cluster 1: (a) CDKN2A, (b) VEGFA, (c) PTGS2, (d) MCM family, and (e) KRT family.

Figure 4. Protein interaction through undirected first neighbor of the candidate genes (highlighted
in yellow) from cluster 1: (a) CDKN2A, (b) VEGFA, (c) PTGS2, (d) MCM family, and (e) KRT family.
Cancers 2023, 15, 853 9 of 13

Table 3. The expression regulation of candidate genes and their functional annotations.

Candidate Gene Functional Annotation

Upregulated Downregulated Term Description
CDKN2A GO:0000079 Regulation of cyclin-dependent protein serine/threonine kinase activity
hsa01522 Endocrine resistance
hsa05166 Human T-cell leukemia virus 1 infection
GO:0008134 Transcription factor binding
hsa05200 Pathways in cancer
hsa04110 Cell cycle
VEGFA PTGS2 hsa04370 VEGF signaling pathway
GO:0071456 Cellular response to hypoxia
hsa05165 Human papillomavirus infection
Negative regulation of cysteine-type endopeptidase activity involved in
GO:0043154
apoptotic process
MCM2, MCM6 MCM4 GO:0006270 DNA replication initiation
hsa03030 DNA replication
GO:0003678 DNA helicase activity
GO:0000727 Double-strand break repair via break-induced replication
KRT1 KRT10 GO:0030280 Structural constituent of epidermis
GO:0018149 Peptide cross-linking
113800 Epidermolytic hyperkeratosis

4. Discussion
Understanding the comprehensive molecular mechanism of cervical cancer patho-
genesis is crucial for developing better screening, management, and treatment for cervical
cancer patients. Thus, many gene-expression studies on cervical cancer via various meth-
ods have been reported [9,25,26]. This systematic review comprehensively explored the
contribution of gene expression and interaction in cervical cancer. From the six selected
studies that reported on the DEGs in cervical samples of cervical cancer patients, 68 genes
were differentially expressed in at least two studies. The PPI network and modular analysis
revealed three significant clusters. All clusters showed high node connection and interac-
tion, in which cluster 1 comprised 20 nodes and 58 edges, cluster 2 consisted of 7 nodes
and 13 edges, and cluster 3 consisted of 4 nodes and 6 edges.
From the functional annotation clustering analysis, the primary molecular mechanism
in cervical cancer is divided into three main mechanisms according to the clusters. The
first mechanism, which includes the DEGs in cluster 1, is primarily involved in the cell
cycle pathway and specifically in regulating the transcription process via ATP binding,
single-stranded DNA binding, and transcription factor binding. The second mechanism,
which includes the DEGs in cluster 2, is mainly involved in cell signaling that regulates cell
proliferation, adhesion, and migration in response to infection and inflammation. The third
mechanism, which includes the DEGs in cluster 3, mainly involves cell differentiation via
the regulation of the keratin filament network through protein heterotetramerization.
The individual genes in each cluster were further analyzed, and seven candidate genes
were selected; these were CDKN2A, VEGFA, PTGS2, MCM2, MCM4, MCM6, KRT1, and
KRT10. The CDKN2A protein is located in the nucleus and cytosol in cells (GO:0005634;
GO:0005829) and is highly expressed in fat, testis, and adrenal tissue [27]. In cervical
tissue, the expression of CDKN2A is typically low, but higher expression (6-fold to 342-
fold change) was reported in cervical cancer tissue in four of the selected studies in this
systematic review [19,21,23,24]. However, this finding is contradictory to an in vitro study
by Luan et al. (2021) that reported on the low expression of CDKN2A in the cervical
cancer cell lines and that the overexpression of CDKN2A inhibits cell proliferation and
invasion of cervical cancer cells by arresting the cell cycle in the G1 phase [28]. Another
contradictory finding was reported in an epigenetics study, in which the cervical cancer
group showed significantly higher CDKN2A methylation than the control group [29],
Cancers 2023, 15, 853 10 of 13

suggesting a low CDKN2A expression level in cervical cancer. Based on the significant
interactions of CDKN2A with transcription factors, signaling molecules, and miRNAs, this
gene has been proposed as a biomarker for cervical cancer prognosis [30]. Thus, due to the
significantly promising role of CDKN2A in cervical cancer, further studies are crucial to
address this conflicting finding.
The VEGFA protein is located in the cytoplasm of a cell and is highly expressed
in thyroid, prostate, lung, and endometrium tissues [27]. In normal cervical tissue, the
expression of VEGFA and its isoform (VEGF165) is low but is upregulated in cervical cancer
tissue [24,31]. VEGFA expression impacts cervical cancer cells’ apoptosis, proliferation,
migration, and invasion [32]. Thus, the inhibition of VEGFA expression could lead to
the inhibition of cell migration and invasive motility in cervical cancer cells [33] while
the PTGS2 (COX2) protein, located in the cytoplasm, is downregulated in cervical cancer
tissue [20,22]. However, the overexpression of PTGS2 has been associated with a poor
prognosis and resistance to cytotoxic therapy [34]. Several signaling mechanisms for
regulating PTGS2 in cervical cancer have been validated, including the EGF and nuclear
factor κB (NF-κB) pathways [35], and PAR2 through an EGFR-dependent mechanism [34].
MCM2, MCM4, and MCM6 are from the minichromosome maintenance complex
(MCM) protein family, which are essential components of the pre-replicative complex for
the initiation of DNA replication and cell division [36]. In cervical cancer, MCM2 [24]
and MCM6 are upregulated while MCM4 is downregulated [20]. However, the higher
expression of MCM2, MCM4, and MCM6 is significantly associated with favorable overall
survival in cervical cancer patients [37].
KRT1 and KRT10 are from the cytokeratin gene family, which is highly expressed in
the upper layer of the epidermis and the endothelial cells [38]. In cervical cancer, KRT1
(type II; “acidic” keratin) and KRT10 (type I; “basic” keratin) are upregulated in cervical
cancer [21]. Thus, it is suggested that KRT1 and KRT10 are involved in the proliferation of
cervical keratinocytes in cervical cancer.
Unlike the MCM family and KRT family, which interact closely within the cluster,
CDKN2A, VEGFA, and PTGS2 show broader interactions when incorporating proteins
from other clusters. Interestingly, the signal transducer and activator of transcription 1
(STAT1) is found to interact with these three genes closely. STAT1 is a tumor suppressor
gene that plays an essential role in apoptotic and anti-apoptotic signaling [39]. From
the functional annotation analysis, CDKN2A, VEGFA, PTGS2, and STAT1 were involved
in cancer pathways (hsa05200; KEGG pathway). The combination and interaction of
these genes in cancer pathogenesis cover the major signaling pathways in cancer, such as
proliferation, angiogenesis, cell cycle, and apoptosis pathways.
Major signaling pathways in proliferation, cell cycle, migration, apoptosis, and DNA
repair are primarily involved in cancer development [40–42]. The phosphatidylinositol
3-kinase/protein kinase B/mammalian target of rapamycin (PI3K/AKT/mTOR) signaling
pathway involved in cell proliferation, survival, invasion, migration, apoptosis, glucose
metabolism, and DNA repair is associated with breast cancer pathogenesis [43]. Mean-
while, the JAK/STAT pathway, PI3K/Akt/mTOR pathway, Ras/Raf/MAPK pathway, and
Wnt/β-catenin pathway contributed to hepatocellular carcinoma by regulating the cell
growth, differentiation, apoptosis, and survival [44].
Besides the major signaling pathways usually exhibited by most cancers, cervical
cancer also exhibits its unique pathways. From the functional annotation clustering per-
formed, cervical cancer was also found to involve pathways related to infections such as
human papillomavirus infection (hsa05165), human cytomegalovirus infection (hsa05163),
human T-cell leukemia virus 1 infection (hsa05166), and staphylococcus aureus infection
(hsa05150). Pathways that respond to infection such as NOD-like receptor-signaling path-
way (hsa04621) and Toll-like receptor-signaling pathway (hsa04620) are also involved in
cervical cancer.
The strength of this systematic review is that it includes studies from various cohorts
and populations. Thus, a more comprehensive molecular mechanism for cervical cancer
Cancers 2023, 15, 853 11 of 13

with reduced effects of genetic variations between populations can be identified. The
homogeneity and bias reduction of the data were assured by only incorporating the DEGs
from the microarray result. The integrated bioinformatic analysis of the DEGs pooled from
the systematic review helps to better interpret the genes’ function and involvement in the
molecular mechanism of cervical cancer.
Nevertheless, the main limitation of this systematic review is the incomplete listing of
differentially expressed genes in some of the studies and the various statistical analyses used
by the included studies in identifying the DEGs. However, the homogeneity of the data was
established by adhering to the strict inclusion criteria and by selecting only DEGs detected
by microarray. This systematic review added new insight into the molecular mechanism of
cervical cancer. It will be beneficial to elucidate further downstream mechanisms of the
DEGs by using RNA interference or gene knockdown strategies.

5. Conclusions
The molecular interactions in the cervical tissue of cervical cancer patients required
extensive investigation to identify a comprehensive molecular mechanism in cervical cancer
pathogenesis. The integrated bioinformatic analysis from this systematic review reveals
three major molecular mechanisms involved in cervical cancer pathogenesis: cell cycle, cell
differentiation, and infection. CDKN2A, VEGFA, PTGS2, MCM2, MCM4, MCM6, KRT1,
KRT10, and STAT1 were identified as key genes that regulate cervical cancer pathogenesis
and progression. These genes can be utilized to develop a biomarker panel for screening
cervical cancer and can become targeted genes for developing better treatment strategies
for cervical cancer.

Supplementary Materials: The following supporting information can be downloaded at https:

//www.mdpi.com/article/10.3390/cancers15030853/s1, Table S1: Quality assessment of the studies
included in the systematic review; Table S2: The common differentially expressed genes in at least
two studies.
Author Contributions: Conceptualization: M.H.E. and N.A.H.; Analysis: M.H.E. and N.A.H.; Writ-
ing: M.H.E., N.A.H. and S.D.; Editing: M.H.E., N.A.H. and S.D. All authors have read and agreed to
the published version of the manuscript.
Funding: This research was funded by the Ministry of Higher Education, Malaysia under Funda-
mental Research Grant Scheme (FRGS/1/2012/SSK01/USIM/03/1; FRGS-PSK-32-50812) and USIM
Internal Grant Scheme (PPPI/FPSK/0118/051000/15418).
Conflicts of Interest: The authors have no competing interests to declare that are relevant to the
content of this article. The funders had no role in the design of the study; in the collection, analyses,
or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

References
1. Arbyn, M.; Weiderpass, E.; Bruni, L. Estimates of incidence and mortality of cervical cancer in 2018: A worldwide analysis. Lancet
Glob. Health 2020, 8, e191–e203. [CrossRef]
2. Fontham, E.T.H.; Wolf, A.M.D.; Church, T.R.; Etzioni, R.; Flowers, C.R.; Herzig, A.; Guerra, C.E.; Oeffinger, K.C.; Shih, Y.-C.T.;
Walter, L.C.; et al. Cervical cancer screening for individuals at average risk: 2020 guideline update from the American Cancer
Society. CA Cancer J. Clin. 2020, 70, 321–346. [CrossRef] [PubMed]
3. Abu-Rustum, N.R.; Yashar, C.M.; Bean, S.; Bradley, K.; Campos, S.M.; Chon, H.S.; Chu, C.; Cohn, D.; Crispens, M.A.; Damast, S.;
et al. NCCN Guidelines Insights: Cervical Cancer, Version 1.2020: Featured Updates to the NCCN Guidelines. J. Natl. Compr.
Cancer Netw. 2020, 18, 660–666. [CrossRef] [PubMed]
4. US Preventive Services Task Force. Screening for Cervical Cancer: US Preventive Services Task Force Recommendation Statement.
JAMA 2018, 320, 674–686. [CrossRef] [PubMed]
5. Huang, J.; Qian, Z.; Gong, Y.; Wang, Y.; Guan, Y.; Han, Y.; Yi, X.; Huang, W.; Ji, L.; Xu, J.; et al. Comprehensive genomic variation
profiling of cervical intraepithelial neoplasia and cervical cancer identifies potential targets for cervical cancer early warning. J.
Med. Genet. 2019, 56, 186. [CrossRef] [PubMed]
6. Balakrishnan, C.K.; Tye, G.J.; Balasubramaniam, S.D.; Kaur, G. CD74 and HLA-DRA in Cervical Carcinogenesis: Potential Targets
for Antitumour Therapy. Medicina 2022, 58, 190. [CrossRef] [PubMed]
Cancers 2023, 15, 853 12 of 13

7. Liu, B.; Zhai, J.; Wang, W.; Liu, T.; Liu, C.; Zhu, X.; Wang, Q.; Tian, W.; Zhang, F. Identification of Tumor Microenvironment and
DNA Methylation-Related Prognostic Signature for Predicting Clinical Outcomes and Therapeutic Responses in Cervical Cancer.
Front. Mol. Biosci. 2022, 9, 872932. [CrossRef]
8. Chen, Q.; Hu, J.; Sun, X.; Long, W. Transcriptome sequencing reveals pathways and genes dysregulated in HPV infected cervical
cancer. Eur. J. Gynaecol. Oncol. 2020, 41, 996–1003. [CrossRef]
9. Balasubramaniam, S.D.; Balakrishnan, V.; Oon, C.E.; Kaur, G. Gene expression profiling of HPV-associated cervical carcinogenesis
in formalin-fixed paraffin-embedded (FFPE) tissues using the NanoString nCounterTM platform. Gene 2022, 825, 146385.
[CrossRef]
10. Hao, Y.; Ye, M.; Chen, X.; Zhao, H.; Hasim, A.; Guo, X. Discovery and validation of FBLN1 and ANT3 as potential biomarkers for
early detection of cervical cancer. Cancer Cell Int. 2021, 21, 125. [CrossRef]
11. Teng, F.; Ruan, H.-J.; Xu, J.; Ni, J.; Qian, B.; Shen, R.; Gao, L.-J. RBBP6 promotes human cervical carcinoma malignancy via JNK
signaling pathway. Biomed. Pharmacother. 2018, 101, 399–405. [CrossRef] [PubMed]
12. Manawapat-Klopfer, A.; Thomsen, L.T.; Martus, P.; Munk, C.; Russ, R.; Gmuender, H.; Frederiksen, K.; Haedicke-Jarboui, J.;
Stubenrauch, F.; Kjaer, S.K.; et al. TMEM45A, SERPINB5 and p16INK4A transcript levels are predictive for development of
high-grade cervical lesions. Am. J. Cancer Res. 2016, 6, 1524–1536. [PubMed]
13. Munn, Z.; Barker, T.H.; Moola, S.; Tufanaru, C.; Stern, C.; McArthur, A.; Stephenson, M.; Aromataris, E. Methodological Quality of
Case Series Studies: An Introduction to the Jbi Critical Appraisal Tool. JBI Evid Synth. 2020, 18, 2127–2133. [CrossRef] [PubMed]
14. Genomics, B.E. Calculate and Draw Custom Venn Diagrams; Bioinformatics & Evolutionary Genomics: Ghent, Belgium, 2022.
15. Szklarczyk, D.; Gable, A.L.; Nastou, K.C.; Lyon, D.; Kirsch, R.; Pyysalo, S.; Doncheva, N.T.; Legeay, M.; Fang, T.; Bork, P.;
et al. The STRING database in 2021: Customizable protein-protein networks, and functional characterization of user-uploaded
gene/measurement sets. Nucleic Acids Res. 2020, 49, D605–D612. [CrossRef]
16. Shannon, P.; Markiel, A.; Ozier, O.; Baliga, N.S.; Wang, J.T.; Ramage, D.; Amin, N.; Schwikowski, B.; Ideker, T. Cytoscape: A
software environment for integrated models of biomolecular interaction networks. Genome Res. 2003, 13, 2498–2504. [CrossRef]
17. Sherman, B.T.; Hao, M.; Qiu, J.; Jiao, X.; Baseler, M.W.; Lane, H.C.; Imamichi, T.; Chang, W. DAVID: A web server for functional
enrichment analysis and functional annotation of gene lists (2021 update). Nucleic Acids Res. 2022, 50, W216–W221. [CrossRef]
18. Kanehisa, M.; Furumichi, M.; Sato, Y.; Kawashima, M.; Ishiguro-Watanabe, M. KEGG for taxonomy-based analysis of pathways
and genomes. Nucleic Acids Res. 2023, 51, D587–D592. [CrossRef]
19. Annapurna, S.D.; Pasumarthi, D.; Pasha, A.; Doneti, R.; Sheela, B.; Botlagunta, M.; Vijaya, L.B.; Pawar, S.C. Identification of
Differentially Expressed Genes in Cervical Cancer Patients by Comparative Transcriptome Analysis. BioMed Res. Int. 2021,
2021, 8810074. [CrossRef]
20. Kim, Y.W.; Bae, S.M.; Kim, Y.W.; Park, D.C.; Lee, K.H.; Liu, H.B.; Kim, I.W.; Jang, C.K.; Ahn, W.S. Target-based molecular signature
characteristics of cervical adenocarcinoma and squamous cell carcinoma. Int. J. Oncol. 2013, 43, 539–547. [CrossRef]
21. Rajkumar, T.; Sabitha, K.; Vijayalakshmi, N.; Shirley, S.; Bose, M.V.; Gopal, G.; Selvaluxmy, G. Identification and validation of
genes involved in cervical tumourigenesis. BMC Cancer 2011, 11, 80. [CrossRef]
22. Miyatake, T.; Ueda, Y.; Nakashima, R.; Yoshino, K.; Kimura, T.; Murata, T.; Nomura, T.; Fujita, M.; Buzard, G.S.; Enomoto, T.
Down-regulation of insulin-like growth factor binding protein-5 (IGFBP-5): Novel marker for cervical carcinogenesis. Int. J.
Cancer 2007, 120, 2068–2077. [CrossRef] [PubMed]
23. Gius, D.; Funk, M.C.; Chuang, E.Y.; Feng, S.; Huettner, P.C.; Nguyen, L.; Bradbury, C.M.; Mishra, M.; Gao, S.; Buttin, B.M.; et al.
Profiling microdissected epithelium and stroma to model genomic signatures for cervical carcinogenesis accommodating for
covariates. Cancer Res. 2007, 67, 7113–7123. [CrossRef]
24. Wong, Y.F.; Cheung, T.H.; Tsao, G.S.W.; Lo, K.W.K.; Yim, S.F.; Wang, V.W.; Heung, M.M.S.; Chan, S.C.S.; Chan, L.K.Y.; Ho, T.W.F.;
et al. Genome-wide gene expression profiling of cervical cancer in Hong Kong women by oligonucleotide microarray. Int. J.
Cancer 2006, 118, 2461–2469. [CrossRef] [PubMed]
25. Chen, T.; Yang, S.; Xu, J.; Lu, W.; Xie, X. Transcriptome Sequencing Profiles of Cervical Cancer Tissues and Siha Cells. Funct. Integr.
Genom. 2020, 20, 211–221. [CrossRef] [PubMed]
26. Øvestad, I.T.; Engesæter, B.; Halle, M.K.; Akbari, S.; Bicskei, B.; Lapin, M.; Austdal, M.; Janssen, E.A.M.; Krakstad, C.; Lillesand,
M.; et al. High-Grade Cervical Intraepithelial Neoplasia (Cin) Associates with Increased Proliferation and Attenuated Immune
Signaling. Int. J. Mol. Sci. 2022, 23, 373. [CrossRef]
27. Fagerberg, L.; Hallström, B.M.; Oksvold, P.; Kampf, C.; Djureinovic, D.; Odeberg, J.; Habuka, M.; Tahmasebpoor, S.; Danielsson,
A.; Edlund, K.; et al. Analysis of the human tissue-specific expression by genome-wide integration of transcriptomics and
antibody-based proteomics. Mol. Cell. Proteom. 2014, 13, 397–406. [CrossRef]
28. Luan, Y.; Zhang, W.; Xie, J.; Mao, J. CDKN2A inhibits cell proliferation and invasion in cervical cancer through LDHA-mediated
AKT/mTOR pathway. Clin. Transl. Oncol. 2021, 23, 222–228. [CrossRef]
29. Truong, P.K.; Lao, T.D.; Le, T.A.H. Cdkn2a Methylation—An Epigenetic Biomarker for Cervical Cancer Risk: A Meta-Analysis.
Pharmacophore 2020, 11, 21–29.
30. Sudha, B.; Poornima, A.; Suganya, K.; Swathi, K.; Kumar, N.S.; Sumathi, S.; Chellapandi, P. Identification of hub genes and role of
CDKN2A as a biomarker in cervical cancer: An in-silico approach. Hum. Gene 2022, 33, 201048. [CrossRef]
31. Patel, K.A.; Patel, B.M.; Thobias, A.R.; Gokani, R.A.; Chhikara, A.B.; Desai, A.D.; Patel, P.S. Overexpression of VEGF165 is
associated with poor prognosis of cervical cancer. J. Obstet. Gynaecol. Res. 2020, 46, 2397–2406. [CrossRef]
Cancers 2023, 15, 853 13 of 13

32. Guo, H.; Li, J.; Fan, F.; Zhou, P. LINC00707 Regulates miR-382-5p/VEGFA Pathway to Enhance Cervical Cancer Progression.
J. Immunol. Res. 2021, 2021, 5524632. [CrossRef] [PubMed]
33. Ying, T.-H.; Lin, C.-L.; Chen, P.-N.; Wu, P.-J.; Liu, C.-J.; Hsieh, Y.-H. Angelol-A exerts anti-metastatic and anti-angiogenic effects
on human cervical carcinoma cells by modulating the phosphorylated-ERK/miR-29a-3p that targets the MMP2/VEGFA axis. Life
Sci. 2022, 296, 120317. [CrossRef] [PubMed]
34. Hugo de Almeida, V.; Guimarães, I.D.S.; Almendra, L.R.; Rondon, A.M.R.; Tilli, T.M.; de Melo, A.C.; Sternberg, C.; Monteiro, R.Q.
Positive crosstalk between EGFR and the TF-PAR2 pathway mediates resistance to cisplatin and poor survival in cervical cancer.
Oncotarget 2018, 9, 30594–30609. [CrossRef] [PubMed]
35. Shi, Y.; Gao, Q.; Liu, Z.; Shen, G.; Sun, X.; Di, X. Identification of Immune and Hypoxia Risk Classifier to Estimate Immune
Microenvironment and Prognosis in Cervical Cancer. J. Oncol. 2022, 2022, 6906380. [CrossRef] [PubMed]
36. Kaur, G.; Balasubramaniam, S.D.; Lee, Y.J.; Balakrishnan, V.; Oon, C.E. Minichromosome Maintenance Complex (MCM) Genes
Profiling and MCM2 Protein Expression in Cervical Cancer Development. Asian Pac. J. Cancer Prev. 2019, 20, 3043–3049. [CrossRef]
37. Wu, B.; Xi, S. Bioinformatics analysis of the transcriptional expression of minichromosome maintenance proteins as potential
indicators of survival in patients with cervical cancer. BMC Cancer 2021, 21, 928. [CrossRef]
38. Han, W.; Hu, C.; Fan, Z.-J.; Shen, G.-L. Transcript levels of keratin 1/5/6/14/15/16/17 as potential prognostic indicators in
melanoma patients. Sci. Rep. 2021, 11, 1023. [CrossRef]
39. Zhang, M.; Liang, L.; He, J.; He, Z.; Yue, C.; Jin, X.; Gao, M.; Xiao, S.; Zhou, Y. Fra-1 Inhibits Cell Growth and the Warburg Effect
in Cervical Cancer Cells via STAT1 Regulation of the p53 Signaling Pathway. Front. Cell Dev. Biol. 2020, 8, 579629. [CrossRef]
40. Niu, G.; Wang, D.; Pei, Y.; Sun, L. Systematic identification of key genes and pathways in the development of invasive cervical
cancer. Gene 2017, 618, 28–41. [CrossRef]
41. Pirim, D. Integrative analyses of molecular pathways and key candidate biomarkers associated with colorectal cancer. Cancer
Biomark 2020, 27, 555–568. [CrossRef]
42. Mollica, V.; Marchetti, A.; Rosellini, M.; Nuvola, G.; Rizzo, A.; Santoni, M.; Cimadamore, A.; Montironi, R.; Massari, F. An Insight
on Novel Molecular Pathways in Metastatic Prostate Cancer: A Focus on DDR, MSI and AKT. Int. J. Mol. Sci. 2021, 22, 13519.
[CrossRef] [PubMed]
43. Miricescu, D.; Totan, A.; Stanescu, S., II; Badoiu, S.C.; Stefani, C.; Greabu, M. PI3K/AKT/mTOR Signaling Pathway in Breast
Cancer: From Molecular Landscape to Clinical Aspects. Int. J. Mol. Sci. 2020, 22, 173. [CrossRef] [PubMed]
44. Alqahtani, A.; Khan, Z.; Alloghbi, A.; Said Ahmed, T.S.; Ashraf, M.; Hammouda, D.M. Hepatocellular Carcinoma: Molecular
Mechanisms and Targeted Therapies. Medicina 2019, 55, 526. [CrossRef] [PubMed]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.