Introduction
In Western countries, colorectal cancer (CRC) is defined as the second most common
reason of cancer deaths and has seen a marked increase in Eastern countries over the
past few decades. There are three CRC types, which include sporadic, familial, and
hereditary.[1] In 25% of cases there is a family history of the disease; yet, most CRCs happen
randomly without such a precedent. There are four types of sporadic CRC: methylator
phenotype of CpG island, hypermutated, raised microsatellite alterations at tetranucleotide
repeats with metastatic behavior, and nonhypermutated.[1] In addition, familial adenomatous polyposis and hereditary nonpolyposis CRC are
two hereditary types of CRC that happen because of an inherited mutation in an autosomal
dominant manner.[2] It has become apparent at the occurrence of the multiphase process of CRC carcinogenesis
happens due to reason of changes in epigenetics and genetic manner progressively accumulating,
causing homeostatic dysregulation, leading to neoplastic transformations.[3]
The role of epigenetic changes in cancer development has only been recently recognized.
In 1942, Conrad H. Waddington used the concept of “epigenetics.” It involves heritable
and potentially reversible changes to the genome's phenotypic expression, which alters
gene expression but does not affect the deoxyribonucleic acid (DNA) primary sequence.
Research has tended to focus on epigenetic modifications which contain noncoding ribonucleic
acid (ncRNA), histone modifications, and DNA methylation alterations that play a part
in CRC pathogenesis.[4] Although screening strategies have improved and more effective treatments are available
for CRC, the prognosis of CRC remains poor. In addition, molecularly targeted agents
have been only reported as active in the case of CRC in metastasis which leads the
economic burden of treatment to be exponentially enhanced. Therefore, identifying
robust biomarkers to diagnose and predict early-stage CRC cases is necessary.[5] Because, unlike genetic modifications, epigenetic alterations can be reversible.
They are a potential new target by which in the development of prevention strategies
for cancer.[2]
[6]
Hypomethylation in CRC
The hypomethylation of DNA significantly play a role in the development of CRC via
activating the transposable elements, removing imprinting, and disrupting the stability
of chromosomes. Global hypomethylation often takes place in repetitive transposable
elements.[7] Aberrant LINE-1 hypomethylation is known to be linked to worse patient survival.[4] An assumption is that inadvertently activating potential proto-oncogenes could induce
the methylation of SINE and LINE sequences to be abnormally hypomethylated, and in
the development of CRC hypomethylated LINE-1 functionally acts. In the promoter region
of CGIs (CpG islands), gene-specific hypomethylation happens in CGIs and causes oncogene
to be abnormally overexpressed.[8]
Hypermethylation in CRC
Advancements in the understanding of CRC molecular pathogenesis, microsatellite instability
(MSI), and chromosomal instability have been shown as primary pathways of molecular
genomic instability, which mainly induces neoplasms. Recent studies have identified
CpG island methylator phenotype (CIMP) as a vital process that contributes to the
development of CRC. In 1999, the phenotype of CIMP was first found in colorectal tumors.[9] The CIMP is affected by numerous genes which play a crucial function within cells.[3] The hypermethylation of CDKN2A and MLH1 genes' promoter region has been reported in most of the CIMP of CRC. Distinct pathological
and molecular properties are found in CIMP-positive tumors, including a tendency for
female gender, mucinous and poor histology, proximal location in the colon, and frequent
mutations (KRAS and BRAF).[10] Suppression of DNA repair genes and tumor suppressor genes including MGMT SEMA3F, APC2, SLC5A8, MLH1, ITGA4, CDKN2A P16, SFRP2, HLTF, and PTCH1 act significantly during progression adenomas to CRC.[11]
Epigenetic Biomarkers for CRC
The National Institutes of Health Biomarkers Definition Working Group described the
expression of either “biomarker” or “biological marker” as a characteristic that is
objectively measured and evaluated as an indicator of normal biological and pathological
processes in 1998. According to current clinical standards, CRC biomarkers are materials
by which people affected with cancer can be identified in an easy and cheap way; and
the prognosis of affected people can be identified without depending on traditional
classifications.[12]
DNA Methylation as Biomarker for Diagnosis
Despite improvements in studies of cancer therapy strategies, premature detection
and removing of precancer lesions remains most successful in CRC and reducing cancer-related
death. Nevertheless, expense, poor patient compliance, and invasiveness place a limitation
on current screening methods, resulting in late diagnosis and therefore unfavorable
prognosis. DNA methylation seems to occur early in tumorigenesis and is therefore
indicated as a possible next generation of cancer diagnostic biomarkers.[13]
[14] The markers of DNA methylation in particular biofluids such as stool and blood (serum
or plasma) have minimum invasiveness properties of diagnosis implemented for the screening
of CRC[4]
[12].
Diagnostic Biomarkers Based in the Blood and Stools
In the 1970s, free DNA biomarkers were discovered by studying abnormal elevations
in serum DNA concentrations among cancer patients.[15] Subsequently, there has been an increased effort to expand into developing diagnostic
tests based on circulating methylation DNA in the blood. There are many genes that
can be utilized as blood-based methylation markers to be able to detect CRC such as
RUNX3, APC, SEPT9, MGMT, WIF1, hMLH1, RASSF2A, HLTF, VIM, ALX4, SFRP2, NGFR, NEUROG1, and TMEFF2, while their range of sensitiveness and specificities are respectively 34 to 90%
and 69 to 100%.[12] The methylated SEPT9 gene is recently identified as a strong potential blood-based methylation biomarker
in the case of distinguishing which samples are cancerous or not via high rate of
specificity and sensitivity (respectively 90 and 72%) compared with other markers.
Thus, the blood-based assay for methylated SEPT9 is marketed by Epi proColon 1.0 (Epigenomics, Seattle, Washington, United States),
ColoVantage (Quest Diagnostics, Madison, New Jersey, United States), and RealTime
mS9 (Abbott Laboratories; Des Plaines, Illinois, United States) and is now marketed
in several countries.[16] The GTP-binding protein is synthesized by SEPT9, and it acts significantly in the
organization of cytoskeletal, division of cells, and remodeling of the membrane. In
addition, many stool-based methylation biomarkers are known to detect CRC earlier,
including hypermethylated HLTF, WIF1, APC, VIM, BMP3, TFPI2, ATM, RASSF2A, SFRP2, NDRG4, CDKN2A, MGMT, GATA4, MLH1, and GSTP1 genes, which were reviewed by several publications.[12]
[16]
The stool-DNA screening is one of the first tests accessible in trading and clinically
confirmed to be used for the analysis of hypermethylated vimentin gene to be able
to detect CRC (ColoSureÔ, Laboratory Corp, Burlington, North Carolina, United States).
The protein of intermediate filament which is crucial in the stabilization of the
cytoskeleton and is highly methylated in CRC is produced by the gene of vimentin (VIM). Respectively, high specificity and sensitivity (90 and 46%) are obtained by this
detection test compared with specificities (95%) and sensitivities (14%) of fecal
occult blood test.[16] The significance and credibility of methylation of DNA in the case of diagnostic
biomarker are approved by scientists, extensive research has recently focused on this
epigenetic alteration in different malignancies, and CRC may be detected earlier using
new methylated genes. For instance, the methylation status of the gene of ADAMTS19 is evaluated on 252 CRC-affected people by Alonso t al; it is concluded as ADAMTS19 is epigenetically silenced in all CRC affected participants due to reason of hypermethylation.
It is added by Alonso et al, that the methylated gene of ADAMTS19 could be used as a new marker for earlier detection of CRC.[17] More research is needed to confirm these genes as diagnostic biomarkers for CRC
in population-based screening.
Methylated DNA as a Treatment Response Predictive Biomarker
It has recently been reported that several aberrantly methylated genes may act as
predictive biomarkers for people with CRC. Most research has not extended beyond the
phase I/II discovery phase. As a predictive marker CIMP has undergone extensive research.
In 2003, people who are affected by CRC and are CIMP-positive gained benefit from
the chemotherapy application of 5-flurouracil (5-FU) without the dependency of survival
and mutation conditions of either TP53 or MSI, which was concluded by Van Rijnsoever
et al.[18] These findings showed that people affected by CRC and in stages II and III with
CIMP-positive tumors benefited from the survival rate after the treatment with 5-FU
were subsequently validated.[19] However, these results were subject to challenge in additional research that studied
a large patient cohort, with the patients with CIMP-positive tumors that did not receive
5-FU treatment surviving longer in comparison to patients with CIMP-negative CRCs.
In the case of comparison between the CIMP-positive CRC patients and patients who
have CIMP-negative tumors, shorter disease-free survival after the treatment of 5-FU
in positive patients has been detected.[20] Even though the differences between the findings may have happened due to reason
of the variety of inheritance or criteria of CIMP between the participants, the utilization
of CIMP as the prognostic marker is promisingly indicated; and for achieving more
information about the interplay between the condition of CIMP and therapeutical reactions
more research is needed to be applied. The conclusions of the credible analysis about
the utilization of genes that are methylated as prognostic markers will give the direction
for more explorations to be applied about the reactions of methylated genes in the
case of prognosis of treatment of CRC.[12]
Histone Modification: A Potential Class of Biomarkers for Colorectal Cancer
The expression of genes and condensation of DNA is controlled by chromatin, while
the nucleosomal core histones are the primary protein constitutive of chromatin. The
nucleosomes consist of an octamer (two copies of each histone) of four core histones
H2A, H2B, H3, and H4 which are wrapped around the DNA. The histone modification happens
in the tails of histones, and histone tails are acetylated, methylated, ubiquitinated,
phosphorylated, and sumoylated. The nucleosomal structure is changed by controlling
of transcription of associated genes and modification of posttranscription, which
leads to chromatin to be either in the form of open and active called euchromatin
or closed and inactive named heterochromatin.[21] For example, acetylation of H3 and H4 and also di- and trimethylation of H3 lysine
4 and H3 lysine 36 (H3K4me2, H3K4me3, H3K36me2, and H3K36me3) are enriched in euchromatin
and result in transcriptional activation. Contrastingly, the transcriptionally inactive
condition is determined via trimethylation of H3 lysine 27 and 9 (H3K27me3 and H3K9me3)
in heterochromatin.[22] After dysregulation of histone modifications in CRC was first discovered,[23] further research revealed that this dysregulation can potentially alter gene expression
patterns in CRC.
The histone modification in the early state of cancers cannot be easily distinguished
due to the technic limitations of analysis in the determination of posttranslational
histone modifications. Therefore, using histone modifications as disease markers is
a subject of conflict. Global modifications of particular histones in the early stage
of tissues have been examined for the generation of disease markers in CRC. According
to research about the H3K9me2, H3K4me2, and H3K9ac modifications detection by immunohistochemical
staining in metastases of the liver, it has been concluded that a high expression
rate of H3K4me2 leads to opposite correlativity withal poor prognosis to be figured
out.[24] In addition, other research into histone modifications in circulating nucleosomes
have shown that there are reduced levels of H3K9me3 and H4K20me3 as possible CRC diagnostic
biomarkers.[25] However, all of these analyses that are studied until now are phase I disease marker
analyses which means they should be under the consideration of “proof of principle.”
Therefore, more research is needed to be applied to determine if these alterations
will be utilized as disease markers in the diagnosis or prognosis of CRC.
Overview of Micro-RNAs
In 1993, micro-RNAs (miRNAs) were identified in the Caenorhabditis elegans as negative posttranscriptional regulators. They are endogenous single-stranded small
RNAs while their length is 18 to 25 nucleotides.[26] The miRNAs bind their target messenger RNAs (mRNAs) and cause target mRNA to be
degraded or the process of translation of mRNA to the protein is inhibited.[27]
miRNAs in Colorectal Carcinogenesis
There are more than 800 articles on alteration in miRNA expression in CRC which has
been published in the database of PubMed.[28] The tumor suppressor genes can be targeted and downregulated by overexpressed miRNAs
in CRC. Furthermore, the proto-oncogenes' activity in the healthy tissues can be downregulated
if miRNAs were silenced in CRC.[29] The CRCs are the process that comes from the adenocarcinoma pathway; Aslam et al
concluded as the upregulated miR135a/b is related to a withal decreased level of APC which causes Wnt signaling pathway to be activated, and normal colonic epithelium
to be transformed into adenoma. Deregulated expression of let-7 miRNA family, miR-18a
and miR-143 (which all have involvement in the regulation of K-RAS expression), of
miR-126 and miR-21 that regulate the phosphatidylinositol-3-kinase (PI-3-K) pathway,
and a cluster of miRNAs (miR-92a, miR-17, miR-19a, miR-18a, miR-19b, and miR-20a)
are linked to regulation of c-myc, which is suspected to have a role in transitioning
from early adenoma to advanced adenoma. Additionally, loss of p53 function is associated
with low expression levels of miR-34a family (a downstream target of p53), which indicates
the role of this miRNA family in transforming adenoma to carcinoma.[30] However, the information about the miRNAs in nontraditional CRC pathways such as
the de novo and serrated pathways is restricted.
MicroRNAs as Clinically Useful Biomarkers for Colorectal Cancer
The miRNA markers can be easily extracted from various clinical sample body tissues
(urine, blood, stool, saliva, etc.) and formalin-fixed paraffin-embedded tissues;
also, they can be stabilized by different laboratory conditions, therefore it has
attracted the interest of scientific researchers for being used as a disease marker
in analyses for more than 5 years.[12]
Use of miRNAs as Diagnostic Biomarkers in CRC
The overexpression of miR-17–3p and miR-92a were identified in CRC.[16] The high expression and secretion of miR-21 in CRC are characterized by the oncogenic
activity of miRNA which makes the miR-21 to be a promising diagnostic marker of CRC
in the early stage of adenoma-carcinoma.[31] The utilization of miR-21 as diagnostic marker of CRC shows high expression and
secretion of miR-21 in the cells of stool or serum samples.[32] Various miRNAs (e.g., miR-155, miR-29a, miR-106a, miR-221, miR-200c, miR-135b, miR-18,
let-7 g, miR-195, etc.) have been defined as diagnostic markers by Okugawa et al.[12] In addition, recent studies have been performed to identify new potency of miRNA
as diagnosis markers (the abnormal expression of miRNAs in the serum, plasma, stool,
or tissues of people who are affected by CRC as new disease marker candidates have
been discovered. Basati et al downregulated two de novo miRNAs, miR-29b and miR-194,
in participants who are affected by CRC and compared with healthy participants and
concluded that miRNAs are powerful CRC serum biomarkers for early diagnosis with high
sensitivity (72 and 80%) and specificity (77 and 75%).[33] According to Peng et al,[34] miR-378 and miR-145 can be used as a diagnostic marker for CRC in the early stages
within high specificities (miR-378 and miR-145, 98 and 60%, respectively) and sensitivities
(100%). It was shown that upregulation of an oncogenic activity of miR-17 -92 cluster
containing miR-92a-2, miR-17, miR-92a-1, and miR- 20a, and downregulation of an oncosuppressive
miR-143 – 145 cluster comprised of miR-143 and miR-145 in tumor tissues and suggested
as potential diagnostic biomarkers for CRC.
Use of miRNAs as a Predictive Biomarker in CRC
Accurately predicting the patient's response to a specific chemotherapeutic drug before
commencing chemotherapy is extremely valuable and subject to comprehensive research.
The drug resistance of miRNAs has been exposed by numerous in vitro-based analyses.
For example, miR-200, miR-10b, miR-215, miR-19b, miR-192, miR-20a, miR-145, miR-21,
miR-140, miR-23a, miR-34, miR-31, and miR-129 family are known to mediate 5-FU[5] resistance.[12]
