JBUON 2021; 26(5): 1723-1734

ISSN: 1107-0625, online ISSN: 2241-6293 • www.jbuon.com

Email: editorial_office@jbuon.com

REVIEW ARTICLE

Basic principles of molecular biology of cancer cell-Molecular

cancer indicators
Emmanuel N. Kontomanolis1, Antonios Koutras2, Athanasios Syllaios3, Dimitrios Schizas3,
Sofia Kalagasidou4, Athanasios Pagkalos5, Dimitra Alatzidou6, Pelagia Kantari1, Thomas
Ntounis2, Zacharias Fasoulakis2
1
Department of Obstetrics and Gynecology, Democritus University of Thrace, Alexandroupolis, Greece. 2Department of Obstetrics
and Gynecology, Alexandra Maternity Hospital, National and Kapodistrian University of Athens, Athens 11524, Greece. 3First
Department of Surgery, National and Kapodistrian University of Athens, Laikon General Hospital, Athens 11527, Greece.
4
Department of Obstetrics and Gynecology, General Hospital Mamatsio of Kozani Greece. 5Department of Obstetrics and
Gynecology, General Hospital of Komotini, Greece. 6Department of Obstetrics and Gynecology, General Hospital of Giannitsa,
Greece.

Summary
Molecular biology of cancer cell is a domain of medical sci- of the sensitivity of cells to repressive signals. The cell has
ence that is rapidly growing in our days. Knowing the ways mechanisms of receiving apoptotic-antitumor signals and
and paths that cancer cells follow is crucial to the preven- mechanisms of execution of these instructions. A percentage
tion of cancer itself. Central role to these paths, concerning of cancers (4-8%) are etiologically linked to germ (stem) cells
the cell cycle and the process of apoptosis, has the protein mutations and occur at an increased frequency in families
p53. The whole mechanism of the cell cycle is activated by (hereditary cancers). Substantial progress in understanding
the action of various mitogens, such as growth factors, hor- the mechanisms of carcinogenesis, filtration and metastasis
mones and cytokines. Carcinogenesis involves alterations of of cancer has highlighted the key role of specific genes, pri-
genes (proto-oncogenes and tumor suppressor genes), which marily oncogenes and tumor suppressor genes.
encode proteins of the signal transduction. Many of the dam-
ages that lead to carcinogenesis may be due to the lack of Key words: molecular biology, cancer cells, molecular cancer
repressive signals for cell division, but also to the absence indicators, principles, cancer

Introduction
Carcinogenesis is a multistage process that eti- tissue, invasion of cancer cells into adjacent tis-
ologically involves mutations in a series of genes sues and, finally, metastasis. Identification of these
that play a role in maintaining the balance between genes and their corresponding gene products, i.e.
cell proliferation and apoptosis, i.e. maintaining a proteins, is of fundamental importance in order to
stable cell mass (number) and also in regulating elucidate the causative pathogenesis of malignant
complex metabolic pathways, which ensure func- transformation, provides additional important in-
tional and structural integrity of cells and tissues formation and is used in the diagnosis and prog-
[1]. Involved disorders in the genes could lead to nosis of cancer and of the potential success of the
uncontrolled cell growth, breakdown of cellular therapeutic treatment [2-4]

Corresponding author: Athanasios Syllaios, MD, MSc. First Department of Surgery, National and Kapodistrian University of
Athens, Laikon General Hospital, Ag. Thoma 17 street, Athens 11527, Greece.
Tel: +30 6972374280; Email: nh_reas@hotmail.com
Received: 30/09/2020; Accepted: 17/11/2020

This work by JBUON is licensed under a Creative Commons Attribution 4.0 International License.
1724 Molecular cancer indicators

The cell cycle and its regulation and DP1 and after being released it could activate
transcription. Phosphorylation of RB is mediated
Most differentiated cells are in a state of rest
by a specific phosphokinase, which is activated af-
(G0), in which they are metabolically active, but
do not duplicate their genetic material nor divide. ter binding to a specific protein, cyclin (which is
However, if they are affected by a mitogenic stim- why phosphokinase is called cdk, cyclin dependent
ulus, the cell cycle is activated and the cells go kinase). There are various cdk kinases and various
through a series of phases, each of which ultimately cyclins (A-E and H). In the specific case of RB acti-
serves to replicate the genetic material and ulti- vation, phosphorylation is initially encoded by the
mately to form two identical daughter cells. These cdk4 / 6 complex with cyclin D and then the cdk2
phases are G1 (Gap1 phase, subdivided into G1a, G1b, complex with cyclin E (Figure 1) [6,8,9].
G1c and G1d), S phase (DNA synthesis phase), G2 Important for the transition from G2 phase to
phase (Gap2) and finally the mitosis phase (M). As M phase is the involvement of MPF (Maturation
long as the mitogen continues to act on the daugh- Promotion Factor), which represents a complex of
ter cells, the cycle is repeated, but if not, the cell cdk1 and cyclin B. This phosphorylates various
will enter in the resting phase G0 and remain there proteins, such as histone H1, lamins and proteins
until the mitogen affects it again. The transition of the mitotic spindle. The affection of the protein
from one phase of the cycle to another is regulated CAK (cdk-activating kinase) is necessary for the
by key molecules, which must ‘remove’ any barriers complete activation of cdk phosphokinases. Cyclins
that inhibit this transition (Figure 1) [5,6]. are involved in other functions such as meiosis,
Such a central key molecule is the phospho- differentiation and apoptosis [8,10].
protein RB, a product of the Rb gene [7]. The pro- Cell cycle regulation also involves cdk kinase
tein in its non-phosphorylated form suppresses inhibitors, the CKI (cyclin dependent kinase inhibi-
the transition from G1 to S phase, which is why tors) belonging to two families, the Cip / Kip family
the corresponding gene is classified in the major (proteins p21, p27, p57) and the Ink4 family (pro-
class of tumor suppressor genes (Table 1). The re- teins p15, p16, p18 and p19). These inhibitors act
pressive effect of the RB is due to the fact that its by forming complexes with kinases [9,11].
non-phosphorylated form binds to and inactivates Central to both regulation of the cell cycle and
transcription factors (E2-E1, DP1). These factors are apoptosis is protein p53, a product of the p53 tumor
involved in the transcription of genes that encode suppressor gene. This tumor suppressor protein
proteins either involved in DNA replication, or are acts in suppressing the transition from G1 phase
structural components of the cell, or are transcrip- to S phase. This is achieved indirectly by acting
tion factors that activate other genes in the cell as a transcriptional activator of the genes, which
cycle. Upon phosphorylation of RB, it breaks down encode the repressive proteins of the cycle (such as
its complex with the transcription factors E2-F1 the p21WAF-1 protein that inactivates cdk kinases

Figure 1. Cell cycle and its regulation.

JBUON 2021; 26(5): 1724

 Molecular cancer indicators 1725

and the MDM2 protein that binds to p53 and in- cell and reaches the nucleus through a ‘’cascade’’
activates it) and Bax and Fas proteins, which have of protein phosphorylations, many of which are
an apoptotic role, and proteins involved in DNA themselves phosphokinases. Eventually, specific
repair. DNA damage, hypoxia, viral cell infiltra- phosphokinases enter the nucleus, phosphorylate
tion and damage due to RB inactivation, lead to and activate a number of proteins involved in the
increased p53 levels, resulting in the inhibition of cell cycle (such as cyclins and transcription factors)
the cell cycle. The purpose of this inhibition is to [10,15,16].
give the cell time to repair DNA damage. If this is Such an example is the signal transduction of
not achieved, the cell enters in the process of apo- growth factors TGFa and PDGF, in which are in-
ptosis (programmed cell death). P53 probably also volved not only the receptors of these factors, but
suppresses the transition from G2 to M. also SOS-Ras-Raf-MEK and MAPK (cytoplasmic)
proteins and fos-jun-myc transcription factors (of
Activation of the cell cycle the core). Signal transduction could also be ini-
The whole mechanism of the cell cycle is ac- tiated by the extracellular matrix, through integ-
tivated by the action of various mitogens, such as rins and finally through the previously mentioned
growth factors (PDGF, EGF, FGF, etc. which are pro- Ras - MAPK cascade. There are many such signal
to-oncogene expression products) [10] hormones transduction systems that are activated by different
and cytokines. Mitogenic signals also originate growth factors and interact [17].
from constituents of the matrix and from intracel- In addition to the stimulatory signals of cell
lular contact molecules (integrins) [12-14]. division, the cell receives several repressive sig-
Proto-oncogene expression products, such as nals. One of the most important is the TGF-β fac-
various transcription factors and cyclins transduct tor, which through activation of its receptor and
mitogenic stimulus from the membrane to the Smad4 cytoplasmic proteins, leads to the activation
nucleus (Table 2). These genes in their mutated of gene transcription and synthesis of the corre-
form (oncogenes) are involved in the process of sponding p15 and p21 proteins. Thus, it inhibits
carcinogenesis. Growth factors trigger the mecha- the formation of complexes between cyclins and
nism of cellular proliferation through interaction cdk, hence the phosphorylation of RB. Additionally,
with membrane receptors, which represent trans- TGF-β suppresses the expression of the c-myc gene,
membrane proteins, with an extracellular domain a transcription factor that is positively involved in
that binds to the growth factor, a transmembrane the regulation of the G1-phase of the cycle [18].
domain and an intracellular domain, which is in-
Disorders of the cell cycle activation-inactivation
volved in the signal transduction through the phos-
mechanisms and of the cycle key components leading
phorylation of a series of proteins. The receptors
to uncontrolled hyperactivity
may either have tyrosine phosphokinase activity
themselves or indirectly activate phosphokinases Carcinogenesis involves alterations of genes
coupled with G-proteins, tyrosine or serine-thre- (proto-oncogenes and tumor suppressor genes),
onine kinases. Whatever the type of receptor is, which encode proteins of the signal transduction
the mitogenic stimulus is transmitted inside the and the cell cycle.

Table 1. The most important tumor suppressor genes/proteins

Genes Protein

p53 53kDa phosphoprotein, transcription factor, induces p21 protein, (cdks

suppressor)

Rb (retinoblastoma) 105kDa phosphoprotein, inactivates (in the non-phosphorylated form) the

transcription factor E2-F1
Wt (Wilm’s tumor) 35kDa protein, contains 4 Zn fingers, transcription factor
BRCA1 (Breast cancer gene 1) 220kDa protein, transcription factor, co-stimulator of p53
BRCA2 (Breast cancer gene 2) Protein 2329 of amino acids, DNA correction
APC (adenomatosis-polyposis colon genes) 300kDa protein, inhibits the formation of a β-catenin complex with
transcription factor Tcf-4 and thus its induction to c-myc
DCC (deleted in colon cancer gene) 190kDa transmembrane protein
NF-1 (neurofibromatosis gene) 250kDa protein, homology with protein which activates GTPase

JBUON 2021; 26(5): 1725

1726 Molecular cancer indicators

Table 2. The most important oncogenes/oncoproteins

Gene Protein

a) Growth factors
C-sis PDGF platelet derived growth factor, b-chain
Int-2 fibroblast growth factor (FGF-3)
hst fibroblast growth factor FGF-4)
b) Growth factor receptors, membrane
c-erbB epidermal growth factor receptor (EGF-R)
HER2/neu (crbB2) receptor structure with tyrosine phosphokinase activity
RET amputated receptor, tyrosine phosphokinase
ros receptor structure
fms mutated receptor CSF-1
c) Phosphokinases, non-receptors, membrane-bound
src
yes
fgr
lck
d) G-proteins, membrane-bound
H-ras GTPase
K-ras GTPase
N-ras GTPase
e) serine-threonine phosphokinases, cytoplasmic
raf/mil
pim-1
cot
mos
f) transcription factors
myc
fos
jun
(Diamandis, 1992, with variations)

Various gene alterations include point muta- Table 3. Various genes / proteins involved in the eti-
tions (such as change of codon 12 [GGC] of the H- opathogenesis of carcinogenesis, cancerous growth, inva-
sion and metastasis (many of which are controlled as bio-
ras gene into GTC in bladder cancer), total or partial logical indicators)
deletion of the gene (loss of heterozygosity, in case
of mutation), the insertion of sequences that can in- Intercellular communication proteins (Cadherins, Type I
activate or activate genes depending on the inser- and II, catenin α and β, integrins), fibronectins
tion site (within the gene or its regulatory parts),
Angiogenic Endothelial Agents (such as the vascular
gene translocation (such as that of c-myc next to endothelial growth factor, VEGF), and its Receptor (VEGF-R),
immunoglobulin genes, resulting in overexpres- FGF and EGF
sion) and gene amplification ie. increase in the
Kallikrein family (gene KLK3 encodes PSA)
number of copies of a gene (such as the c-erbB-2
in breast and ovary cancer) (Table 3) [15,19-21]. Other proteases (MMP-II, MMP-9, MMP-2, which break
down the matrix and collagen), cathepsins
These alterations are partially reversible
thanks to the existence of a DNA repair mecha- Telomerase
nism involving the hMS1, hMS2, hPMS1 and Genes encoding repair enzymes (mismatch repair genes)
hPMS2 repair genes and their corresponding re- hMSH1, hMSH2, hPMS1, hPMSH2.
pair enzymes that they encode. However, these Genes and Proteins Positive or Negative in Apoptosis
genes can also mutate and be inactivated, thus (Positive: Bcl, BclXL, MC-1. Negative: Bax, Bad, BcX5)

JBUON 2021; 26(5): 1726

 Molecular cancer indicators 1727

removing the repair capacity of the cell and pro- lead to activation of these oncoproteins are often
moting carcinogenesis [22]. observed in various types of cancers [30,31].
Mutations in oncogenes and tumor suppressor
genes make cell proliferation being independent of The role of telomeres and telomerase in carcinogenesis
extracellular mitogenic signals. Some cells (glio- Hayflick’s work has shown that cells in culture
blastomas and sarcomas) produce growth factors, have limited cell division capacity, ranging from
such as PDGF and TGFa, by themselves, thus creat- 60-70 divisions for most cell types. It seems that
ing a permanent autocrine cycle [23,24]. the telomeres-telomerase system plays a role in
Another way of self-regulation is the disrup- limiting the reproductive process [24].
tion of growth factor receptors, due to either over- The ends of chromosomes consist of thousands
expression or mutations. In the first case, the cell of repeated sequences of 5-6 nucleotides, the tel-
becomes sensitive to concentrations of growth omeres, which exert protective action on the chro-
factors that would not normally lead to mitosis: if mosomes. At each cell division, 50-100 nucleotide
the production of receptors is excessive, it could sequences are lost from the ends of these telom-
lead to mitosis and absence of growth factors. In eres. Thus, after a certain number of cell divisions,
the latter case, the receptor is transformed into a the telomeres lose their ability to exert protective
constitutively active molecule, activating the sig- action on the ends of chromosomes, which are then
nal transduction process in the absence of growth closely coalesced, causing karyotype disorders and
factor. Spontaneous mitogenic stimuli may also eventually the cell death [32]. Cancer cells show
result from disorders of cellular interconnection
increased expression of the telomerase enzyme
molecules, such as integrins and beta-catenin [25].
which adds repetitive hexanucleotides to the ends
Many of the molecules involved in signal
of the DNA and restores telomere length and cell
transduction are involved in the carcinogenic pro-
division ability. The importance of telomeres and
cess in their mutated form. Typical examples are
telomerase is reinforced by the fact that expres-
the Ras mutant protein, which activates the next
sion of telomerase in cell transfection experiments
steps of the signal transduction cascade without
renders them immortal. Similar results have been
itself being previously stimulated by mitogens. It
obtained with transgenic animals [33].
is possible that some of the disorders leading to
the induction of spontaneous mitogenic signals Apoptosis
originate from the layer’s cells [26].
Many of the damages that lead to carcinogen- As stated in the Introduction, the rate of apo-
esis may be due to the lack of repressive signals for ptosis (programmed cell death) plays a key role
cell division, but also to the absence of the sensitiv- in maintaining a stable cell number. The cell has
ity of cells to repressive signals. Damages related to mechanisms of receiving apoptotic-antitumor sig-
TGFβ, which normally suppresses cell division, are nals and mechanisms of execution of these instruc-
well-studied and involve either its downregulation, tions. Insulin-like growth factors IGF-R and IL-3R
or its mutation leading to function loss [27]. Moreo- act in an anti-apoptotic way through the IGF-R and
ver, the absence of functionality of some integrins, IL-3R receptors, respectively, whereas FAS and
which normally also send repressive signals, con- TNFα act in an apoptotic way through the FAS TNF-
tributes to the cancer phenotype. The mutation of R1 receptors, respectively [5]. Intracellular mecha-
proteins, that transmit repressive stimuli from the nisms determine whether or not the cell is func-
cell membrane to the nucleus, has a similar effect tioning properly and, in the event of DNA damage,
(eg mutation of the Smad4 protein) [28]. oncogenic activity, survival factors deficiency or hy-
Disorders in cell cycle components, which are poxia, activate the apoptotic pathway by releasing
located in the cell nucleus, such as RB protein, the mitochondrial cytochrome C, which activates
cyclins, cdk kinases and transcription factors are a series of proteases, the caspases, which eventu-
significant and well-studied. Mutations in the Rb ally, selectively disrupt subcellular structures, or-
gene, observed in many cancers (colon, small cell ganelles and genome [34,35]. Many proteins act on
lung, esophagus, breast, prostate) (Table 3), abolish apoptosis via mitochondria and the release of cyto-
the ability of the molecule to form a complex with chrome C, such as Bax, Bak, Bid, Bim (apoptotic) and
the transcription factor E2F-1, which is free to act Bcl-2, B cl-XL, Ccl-W (anti-apoptotic). In the event
constitutively [7,29]. of DNA damage and its non-repair, p53 promotes
Mutations of CDK-41 in its interaction with the apoptosis by enhancing Bax expression. In contrast,
p15 INK4B suppressor protein may render it un- overexpression of bcl-2, as in the translocation of
able to accept repressive signals. Many mutations the gene to a strong transcription site, acts antago-
in transcription factor genes (fos, myc, etc.) that nistically and positively on oncogenesis [36].

JBUON 2021; 26(5): 1727

1728 Molecular cancer indicators

Resistance to apoptosis is achieved in several applications in various types of cancers, leading to

ways. One common way is to mutate p53 and its reduced tumor growth and shrinkage [24].
corresponding protein (observed in 50% of can-
cers). The anti-apoptotic mechanism could also be Carcinogenesis as a multistage process
activated by extracellular factors (IGF-1/2, IL-3), Many of the aforementioned genes are in-
by signals from Ras, or by loss of the pTEN tumor volved in the causative pathogenesis of malignant
suppressor gene. Another mode of anti-apoptotic transformation and the manifestation of the can-
action is through a mutant FAS death factor re- cer phenotype. In some cancers, it appears that
ceptor, such as in colon and lung cancer, which in specific mutated genes are involved in a defined
the mutant form does not understand the apoptotic timing. The case of colon cancer is characteristic,
messages of FAS. It is very likely that all cancer where the conversion of normal intestinal epithe-
cells have mutations that allow them to bypass lium, through intermediate stages (adenomas), into
apoptosis [37,38]. metastatic tumor cell involves activation of onco-
genes and inactivation of tumor suppressor genes,
Genes involved in tumor perfusion, invasive process in a particular sequence (FAP - Ras - DCC - p53)
and cancer metastasis [39]. Likewise, in the case of endometrial cancer,
The first phase of carcinogenesis is followed the conversion of normal cell, through a stage of
by an increase in cell mass, tumor cell invasion hyperplasia, into cancer and cancer-metastatic cell
into adjacent tissues and vessels and metastasis involves sequential activation of the ras oncogene,
(migration-installation into distant tissues and or- inactivation of the p53, DCC and Rb oncogene-sup-
gans). Cell migration is based on invasion of the pressor genes and finally activation of oncogenes
cytoplasmic membrane on the guiding side of the of erbB2, myc and Sms. In other cases, it appears
cell, formation of new extracellular adhesion sites that gene alteration is accidental and only the cu-
on the guiding part of the cell, release of adhesion mulative sum of the gene lesions, irrespective of
sites at the back, and contraction of the cytoskeletal their timing, is significant. This explains the high
elements [1]. incidence of cancer in old age [40,41].
A number of genes are activated to accomplish
these processes, encoding, inter alia, autocrine mo- The utilisation of genes and their expression products
involved in the etiology of cancer as biological cancer
tility factors, angiogenesis factors and their recep-
indicators
tors (e.g. VEGF, VEGF-R), proteases (collagenases,
kallikreins), cell adhesion and interconnection Progress in elucidating the mechanisms of
molecules and growth factors (Table 4). The im- carcinogenesis and identifying the involvement
portance of these genes in the above processes is of many genes (such as oncogenes and tumor sup-
demonstrated in their neutralization experiments pressor genes) and their expression products in the
and their expression products (with antisense carcinogenic process, as well as the introduction of
RNAs, with antibodies, etc.), and from pilot clinic sensitive molecular detection techniques of slight

Table 4. Frequency (%) of oncogene activation or tumor suppressor gene inactivation in adult cancers

Cancer Oncogenes Tumor suppressor genes

Her2/neu ras myc p53 Rb APC DCC

Exocrine pancreas cancer 10-20 75-90 40 20 30 50
Colon, colorectal cancer 10-20 65 5-65 70-75 35 70-80 70-80
Small cell lung cancer rarely 0 25 99-100 95-100 - -
NSCLC* 55-60 40(adeno) 48 50 10-20 30 15
Ovarian cancer 30 20-25 1-40 50 10 rarely 35
(LOH)**
Breast cancer 25-40 rarely 20-30 25-50 20 5-10 30
(LOH)**
Prostate cancer 70-80 0-5 50 10-20 25 20 25
Oesophageal cancer 20 rarely 5-10 33-50 35-50 5-15 5-10
Gastric cancer 20 20 5-20 20-50 rarely 20 50
*NSCLC = non small cell lung cancer, **LOH = loss of heterozygosity, (from Lyerly and Sullenger, 1994)

JBUON 2021; 26(5): 1728

 Molecular cancer indicators 1729

quantities of DNA, RNA or protein, have been de- As of late, there were different types of micro-
terminants for the utilization of the aforementioned RNAs described as potential biomarkers used for
biological cancer indicators genes / proteins [24]. An screening cancer (Figure 2). This means that from
ideal cancer indicator is one that ensures early diag- the moment that the levels of distinct microRNAs
nosis, has predictive value for disease progression are abnormal, they can be observed at the begin-
and detection of recurrence, has predictive value for ning of cancer, during its progression and, also, af-
treatment, is suitable for population control, is read- ter metastasis. Moreover, they are less or non-inva-
ily measurable and easy to use and is not costly. sive and can be obtained by liquid biopsies (i.e. via
None of the indicators used today have all of these urine, saliva, semen, or breast milk) without caus-
properties, a disadvantage bypassed with the utiliza- ing severe damage and pain to people [46]. Addi-
tion of more than one indicator. It should be empha- tionally, when the pattern of circulating microRNAs
sized that genes/proteins may be defective in differ- becomes dynamically expressed, it can be associ-
ent types of cancer, but damage to specific genes is ated with tumor progression and aggressiveness.
characteristic of some forms of cancer [15,42]. On the other hand, the differences of the po-
The methodology for detecting biological indi- tential microRNA markers between healthy people
cators allows the detection of molecular damage at and cancer patients are, usually, tiny. For instance,
the level of DNA, RNA and protein. Τhe method of in blood sampling methods there should be a care-
qualitative and quantitative polymerase chain reac- ful microRNA consideration, as their levels do not
tion is important (PCR), combined with mutation de- differ significantly between artery and vein. Gen-
tection techniques (such as single stranded spatial erally, microRNAs can be either clearly related to
polymorphism (SSCP), restriction fragment length cancer (oncomiR) – in a certain cancer type – or a
polymorphism (RFLP), etc.) and DNA sequencing. At suppressor in another [45].
the detection level of the protein, immunochemical To be more specific, for instance regarding breast
methodology using polyclonal or monoclonal an- cancer, there are two miRNAs miR-155 and miR-195
tibodies in the form of western blots provides sig- that were reported in studies as potential biomark-
nificant sensitivity. These techniques could also be ers, but even them are not still importantly altered
applied at the cell/histological level, on fine-needle in order to be fully used in an early detection of
biopsies or even on circulating cellular elements breast cancer; they are indicative, though [45,47,48].
[43,44]. Moreover, as far as the ovarian cancer is concerned,
According to Wang et al microRNAs are en- the miRNAs highly expressed in patients suffering
dogenous single-stranded non-coding small RNA from ovarian cancer are serum miR-21, miR-29b,
molecules that can be secreted into the circulation miR-92, miR-93 and miR-126, while miR-21, miR-92
and exist stably. They usually exhibit aberrant ex- and miR-93 have been indicated before CA-125 (the
pression under different physiological and patho- only diagnostic index so far indicating ovarian can-
logical conditions [45]. cer) started to increase; this showcased that those

Figure 2. Differentiated-expressed circulating miRNA in NSCLC patients.

JBUON 2021; 26(5): 1729

1730 Molecular cancer indicators

three miRNA types could be used as biomarkers for In invasive porcine cancer, the HER-2 / neu
an early ovarian cancer diagnosis [49-51]. gene is overexpressed in 15% of the cases, and
All in all, it is important to implement micro- in invasive non porcine (in situ) in 56%. Gene
arrays, which allow a simultaneous evaluation overexpression is associated with age, grade of
of the expression of thousands of genes and, on differentiation (higher expression in low-grade
this basis, a possible classification of tumors into differentiation of cancer and lower expression in
categories for their aggressiveness or response to high-grade), absence of estrogen and progesterone
treatment. Proteomic, or simultaneous analysis of receptors, increased expression of cathepsin D, p53
many proteins with potential prognostic or pre- overexpression, tumor size, lymph node invasion
dictive importance, will have a wide application and ploidy (higher expression in tetraploid cells
[52]. Of the many genes / proteins studied, only compared to diploid ones). Also, overexpression of
a few have proven to be important biological in- HER-2 / neu is associated with a poor response to
dicators in sporadic cancers. On the contrary, the endocrine therapy. The results are contradictory
contribution of specific genes as indicators in cases as far as chemotherapy is concerned [60,61]. The
of hereditary cancers is valuable (see the section therapeutic potential of HER-2 / neu overexpres-
‘Biological Indicators in Hereditary Cancer’). Table sion, using monoclonal antibodies instead of pro-
3 shows the frequency of activation of oncogenes tein (Herceptin), is promising.
and inactivation of tumor suppressor genes in cer- C-myc expansion is observed in approximately
tain adult cancers [24]. 1/5 of breast cancers, whereas in patients with me-
In lung adenocarcinoma, K-ras mutation is a tastases the expansion is found in 1/3 of the cases
prognostic indicator of malignancy. In almost all cas- and has been associated with early relapses and a
es of small cell carcinomas, inactivation of p53 and short survival period, especially in tumors negative
Rb tumor suppressor genes is observed [53]. In 27% for estrogen receptors without lymph node invasion.
of non-small cell cancers (NSCLC), overexpression of The myc gene is an independent prognostic indica-
the mos oncogene (encoding threonine-serine phos- tor of reduced survival and characterises expansive
phokinase) is observed. The expression is greater in tumors. Its predictive value for endocrine therapy is
stages II and III (34%) than in stage I (17%), without insignificant, whereas a better response to chemo-
expansion of the gene. There is a correlation between therapy has been observed in tumors without gene
increased gene expression and decreased apoptosis. expansion than in those with expansion [62-64].
Lesions of p53 were found in 86% of tumors [54]. P53 is the most frequently mutated gene (50%
Pancreatic cancer has a high incidence of ras of the cases) in breast cancer. The detection of onco-
oncogene activation and APC tumor suppressor protein in the nucleus is almost always evidence of a
gene inactivation [55]. mutant gene leading to an increase in the half-life of
In colon cancer, inactivation of p53, APC and the protein. However, some mutations do not affect
DCC tumor suppressor genes and activation of ras the half-life of the protein, so analysis at the gene
and myc oncogenes are characteristic [56]. level is necessary. Mutations may concern one or
In ovarian cancer, increased expression of both alleles, with or without loss of heterozygosity.
EGFR and expansion and overexpression of Her-2 Mutations of p53, mainly at some conserved sites,
/ neu (c-erbB2) are poor prognostic indicators. In are associated with a poor prognosis and are an in-
cases of c-erbB2 overexpression, the oncoprotein dependent prognostic indicator and go along with
is also detected in the serum in equivalent quanti- the absence of estrogen / progesterone receptors, a
ties with CAI25 indicator. Colony-activating factor greater grade of differentiation (III> I) and increased
(CSF-1) is also detected in patient serum and the cell division. In terms of their predictive importance,
tumor overproduces mutant c-fms (CSF-1 recep- endocrine therapy (tamoxifen administration) is less
tor). Overexpression of K1-ras and expansion and valuable in patients with mutations in p53, than in
overexpression of c-myc is also observed in about those without mutations. In contrast, chemotherapy
25% of the cases [57,58]. resulted in increased survival in p53-positive patients,
Molecular biological indicators have proved although this observation is also disputed [65,66].
very useful in both sporadic and hereditary breast Overexpression of bcl-2 has been associated
cancers. The proto-oncogene HER-2 / neu and with a good prognosis, whereas low expression of
the corresponding oncoprotein have been found bax with a poor one.
in non-invasive breast cancer, at 5-55% (average In prostate cancer, many gene disorders are ob-
26%). There is gene expansion in 1/3 of cancers, served, probably in a specific order, which charac-
which could be detected with Southern blot, with terizes the stage of the carcinogenic process. Thus,
PCR, as well as with in situ hybridization with fluo- cells in the pre-cancerous stage (carcinogenic initi-
rescence techniques (FISH) [53,59]. ation stage) show c-myc expansion, overexpression

JBUON 2021; 26(5): 1730

 Molecular cancer indicators 1731

Table 5. Genes involved in hereditary cancer The most important hereditary cancers include
breast cancer, retinoblastoma, Wilms tumor, thy-
Gene Disease roid cancer and hereditary non-polyposis colorec-
Rb Retinoblasoma tal cancer (Lynch syndrome, HNPCC). Table 5 lists
APC Familial adenomatous polyphagia (FAP) the genes whose mutations characterize specific
hMSH2, hPMS2 Hereditary nonpolyposis colorectal
hereditary cancers. Detection of the mutations in
cancer (HNPCC) these cases is crucial in monitoring individuals and
taking precautions and early therapeutic measures
BRCAI, II Breast cancer, Ovarian cancer
(such as breast or thyroid removal, etc.) [29,62].
RET Multiple endocrine type 2
Hereditary breast cancer accounts for 1/4 of all
breast cancer cases diagnosed before the age of
thirty [31]. In 45% of hereditary cancers, there is
of bcl-2, c-erbB2 and EGFR and, in 25% of the cases, a mutation in the BRCA1 tumor suppressor gene
telomerase activation. Activation of telomerase (in and in the other 45% there is a mutation of the
75% of cases), ras activation and loss of E-cadherin BRCA2 gene [20,74,75]. Both genes are stimulated
are observed in cancer cells at the promotion stage, by estrogens. Their expression products (Table 1)
whereas invasive metastatic cells overexpress the act as transcription factors, and BRCAI suppresses
mutant tumor suppressor gene p53 and the angio- estrogen-induced transcription. Mutations, which
genic factor FGF. Overexpression of p53 is associ- affect different areas of the genes, lead to non-
ated with aggression, risk of relapse and a high functional proteins or proteins that inhibit the ac-
degree of Gleason [67-69]. tion of the corresponding physiological proteins
The expansion of N-myc into the neuroblas- (dominant mechanism of action). The predisposi-
toma, the fourth most common pediatric tumor, is tion to breast cancer is caused by the existence
characteristic. The degree of the expansion is asso- of the mutated allele, while an upcoming loss of
ciated with tumor aggression and is an indication the normal allele (loss of heterozygosity) leads to
for a more aggressive treatment [70,71]. carcinogenesis [75,76].
To sum up, all cancer types have their own ex-
pression and mechanism and their own biomarkers Conclusion
that can be used as tools to an early diagnosis and
monitoring of the disease course. Here is where the Substantial progress in understanding the
need for better proteomics understanding emerges mechanisms of carcinogenesis, filtration and me-
as it could be extremely helpful to the whole issue tastasis of cancer has highlighted the key role of
of cancer comprehension and treatment. specific genes, primarily oncogenes and tumor
Proteomics is the use of quantitative protein- suppressor genes and their expression products in
level measurements of gene expression to charac- these processes and led to the effort of using them
terize biological processes (e.g., disease processes as biological indicators.
and drug effects) and decipher the mechanisms of In the case of hereditary cancers, the clinical
gene expression control [52,72]. use of molecular biological indicators has proven
Proteomics – or in this case – oncoproteomics to be important and in many cases life-saving. In
are proteins related to cancer cell pathogenesis un- sporadic cancers their benefits mainly focus on
derstanding. They help clinical practice, including their prognostic and predictive importance. It is
diagnosis as well as screening via the mechanism of important to apply the microarrays method in this
proteins in order to assist early cancer prognosis [52]. field, where thanks to the sequencing of the human
genome and its consequent ability to simultane-
Biological indicators in hereditary cancer ously analyze the expression of thousands of genes,
A percentage of cancers (4-8%) are etiologically it would be possible to classify cancers in different
linked to germ (stem) cells mutations and occur at behavioral groups based on their gene expression
an increased frequency in families (hereditary can- patterns and the association of these patterns with
cers). Sufferers inherit a mutated tumor suppressor possible prognosis and prediction. Significant de-
gene: however, the tumor suppressor protein pro- velopments are also expected regarding the ability
duced by the healthy allele prevails and, thus, the of gene analysis in individual circulating (pre) can-
phenotype appears normal. However, during the cer cells to timely detect a potential cancer process.
life of a person, the loss of the other allele occurs
(LOH, loss of heterozygocity) and consequently the Conflict of interests
loss of its normal product expression, which results
in the development of cancer [23,53,73]. The authors declare no conflict of interests.

JBUON 2021; 26(5): 1731

1732 Molecular cancer indicators

References
1. Derelanko MJ. Carcinogenesis. Handb Toxicol (Second 19. Yu Z, Wang C, Wang M, Li Z et al. A cyclin D1/microR-
Ed), 2001. doi:10.5694/j.1326-5377.1984.tb103974.x. NA 17/20 regulatory feedback loop in control of breast
2. Wang RA, Li ZS, Zhang HZ et al. Invasive cancers are cancer cell proliferation. J Cell Biol 2008. doi:10.1083/
not necessarily from preformed in situ tumours - an jcb.200801079.
alternative way of carcinogenesis from misplaced 20. Holschneider CH, Berek JS. Ovarian cancer: Epidemiol-
stem cells. J Cell Mol Med 2013;17:921-6. doi:10.1111/ ogy, biology, and prognostic factors. Semin Surg Oncol
jcmm.12078. 2000. doi:10.1002/1098-2388(200007/08)19:1<3::AID-
3. Balkwill FR, Capasso M, Hagemann T. The tumor micro- SSU2>3.0.CO;2-S.
environment at a glance. J Cell Sci 2012. doi:10.1242/ 21. Stuckey A. Breast cancer: Epidemiology and risk fac-
jcs.116392. tors. Clin Obstet Gynecol 2011;54:96-102. doi:10.1097/
4. Society AC. Cancer Facts & Figures 2016. Cancer Facts GRF.0b013e3182080056.
Fig 2016 2016. 22. Lieber MR. The Mechanism of Double-Strand DNA
Break Repair by the Nonhomologous DNA End-Join-
5. Yang N, Sheridan AM. Cell Cycle. Encycl. Toxicol (Third
ing Pathway. Annu Rev Biochem 2010. doi:10.1146/
Ed), 2014. doi:10.1016/B978-0-12-386454-3.00273-6.
annurev.biochem.052308.093131.
6. Weitzman MD, Wang JYJ, Verma V. Cell cycle: DNA
23. Nenclares P, Harrington KJ. The biology of cancer.
damage checkpoints. Encyclopedia of Biological chem-
Medicine (United Kingdom) 2020. doi:10.1016/j.
istry III (3rd Edition). Elsevier, 2021, 2-9.
mpmed.2019.11.001.
7. Giacinti C, Giordano A. RB and cell cycle progression.
24. Gariglio P. Oncogenes and tumor suppressor genes.
Oncogene 2006. doi:10.1038/sj.onc.1209615.
Mol Oncol Princ Recent Adv, 2012. doi:10.2174/97816
8. Lim S, Kaldis P. Cdks, cyclins and CKIs: roles beyond 0805016111201010064.
cell cycle regulation. Development 2013. doi:10.1242/
25. Komiya Y, Habas R. Wnt signal transduction pathways.
dev.091744.
Organogenesis 2008. doi:10.4161/org.4.2.5851.
9. Hochegger H, Takeda S, Hunt T. Cyclin-dependent ki- 26. Rampias T, Vgenopoulou P, Avgeris M et al. A new tu-
nases and cell-cycle transitions: Does one fit all? Nat mor suppressor role for the Notch pathway in bladder
Rev Mol Cell Biol 2008. doi:10.1038/nrm2510. cancer. Nat Med 2014. doi:10.1038/nm.3678.
10. Salazar-Roa M, Malumbres M. Fueling the Cell Di- 27. Lewis AM, Varghese S, Xu H, Alexander HR. Interleu-
vision Cycle. Trends Cell Biol 2017. doi:10.1016/j. kin-1 and cancer progression: The emerging role of
tcb.2016.08.009. interleukin-1 receptor antagonist as a novel therapeu-
11. Kim YK, Yu J, Han TS et al. Functional links between tic agent in cancer treatment. J Transl Med 2006;4.
clustered microRNAs: Suppression of cell-cycle inhibi- doi:10.1186/1479-5876-4-48.
tors by microRNA clusters in gastric cancer. Nucleic 28. Hodivala-Dilke KM, Reynolds AR, Reynolds LE. In-
Acids Res 2009. doi:10.1093/nar/gkp002. tegrins in angiogenesis: Multitalented molecules
12. Gavalas NG, Liontos M, Trachana SP et al. Angiogen- in a balancing act. Cell Tissue Res 2003;314:131-44.
esis-related pathways in the pathogenesis of ovarian doi:10.1007/s00441-003-0774-5.
cancer. Int J Mol Sci 2013. doi:10.3390/ijms140815885. 29. Nevins JR. The Rb/E2F pathway and cancer. Hum Mol
13. Sherwood LM, Parris EE, Folkman J. Tumor Angiogen- Genet 2002. doi:10.1093/hmg/10.7.699.
esis: Therapeutic Implications. N Engl J Med 1971. 30. Lal A, Kim HH, Abdelmohsen K et al. p16INK4atransla-
doi:10.1056/NEJM197111182852108. tion suppressed by miR-24. PLoS One 2008. doi:10.1371/
14. Joimel U, Gest C, Soria J et al. Stimulation of angiogen- journal.pone.0001864.
esis resulting from cooperation between macrophages 31. Thillainadesan G, Chitilian JM, Isovic M et al. TGF-β-
and MDA-MB-231 breast cancer cells: Proposed mo- Dependent Active Demethylation and Expression of
lecular mechanism and effect of tetrathiomolybdate. the p15 ink4b Tumor Suppressor Are Impaired by the
BMC Cancer 2010;10. doi:10.1186/1471-2407-10-375. ZNF217/CoREST Complex. Mol Cell 2012. doi:10.1016/j.
15. Malumbres M, Barbacid M. To cycle or not to cycle: molcel.2012.03.027.
a critical decision in cancer. Nat Rev Cancer 2001. 32. Chen JJL, Podlevsky JD. Telomeres and Telomerase.
doi:10.1038/35106065. Encycl Cell Biol, 2016. doi:10.1016/B978-0-12-394447-
16. He T, Qi F, Jia L et al. Tumor cell-secreted angiogenin 4.10042-2.
induces angiogenic activity of endothelial cells by sup- 33. Artandi SE, DePinho RA. Telomeres and telomerase
pressing miR-542-3p. Cancer Lett 2015;368:115-25. in cancer. Carcinogenesis 2009. doi:10.1093/carcin/
doi:10.1016/j.canlet.2015.07.036. bgp268.
17. Zeng Q, Li S, Chepeha DB et al. Crosstalk between tu- 34. Chen G, Goeddel D V. TNF-R1 signaling: A beauti-
mor and endothelial cells promotes tumor angiogen- ful pathway. Science (80- ) 2002. doi:10.1126/sci-
esis by MAPK activation of Notch signaling. Cancer ence.1071924.
Cell 2005. doi:10.1016/j.ccr.2005.06.004. 35. Gemma A, Takenaka K, Hosoya Y et al. Altered expres-
18. Ohnuki H, Jiang K, Wang D et al. Tumor-infiltrating sion of several genes in highly metastatic subpopu-
myeloid cells activate Dll4/Notch/TGF-β signaling lations of a human pulmonary adenocarcinoma cell
to drive malignant progression. Cancer Res 2014. line. Eur J Cancer 2001;37:1554-61. doi:10.1016/S0959-
doi:10.1158/0008-5472.CAN-13-3118. 8049(01)00154-X.

JBUON 2021; 26(5): 1732

 Molecular cancer indicators 1733

36. Ren D, Tu HC, Kim H et al. BID, BIM, and PUMA are 54. Erenpreisa J, Cragg MS. MOS, aneuploidy and the ploidy
essential for activation of the BAX- and BAK-dependent cycle of cancer cells. Oncogene 2010. doi:10.1038/
cell death program. Science (80- ) 2010. doi:10.1126/ onc.2010.310.
science.1190217. 55. Dindyal S, Spalding D. Pancreatic cancer. Medi-
37. Gulbins E, Bissonnette R, Mahboubi A et al. FAS-induced cine (United Kingdom) 2019. doi:10.1016/j.
apoptosis is mediated via a ceramide-initiated RAS mpmed.2019.04.008.
signaling pathway. Immunity 1995. doi:10.1016/1074- 56. Muzny DM, Bainbridge MN, Chang K et al. Compre-
7613(95)90142-6. hensive molecular characterization of human colon and
38. Cui M, Zhang Q, Qiu X, Jin F. The role of IL-11 and rectal cancer. Nature 2012. doi:10.1038/nature11252.
IL-Ra in angiogenesis of breast cancer. Int J Clin Exp 57. Kamura T, Jeon JD. Lymph Node Metastasis in a Gy-
Pathol 2016;9:11682-7. necologic Malignancy. Yonsei Med J 2002;43:783.
39. Joyce JA, Pollard JW. Microenvironmental regulation of doi:10.3349/ymj.2002.43.6.783.
metastasis. Nat Rev Cancer 2009. doi:10.1038/nrc2618. 58. Lengyel E. Ovarian cancer development and metastasis.
40. Abel EL, DiGiovanni J. Multistage carcinogenesis. Curr Am J Pathol 2010. doi:10.2353/ajpath.2010.100105.
Cancer Res 2011. doi:10.1007/978-1-61737-995-6_2. 59. Lee SK, Bae SY, Lee JH et al. Distinguishing low-risk
41. Barcellos-Hoff MH, Lyden D, Wang TC. The evolution luminal a breast cancer subtypes with Ki-67 and p53
of the cancer niche during multistage carcinogenesis. is more predictive of long-term survival. PLoS One
Nat Rev Cancer 2013. doi:10.1038/nrc3536. 2015;10. doi:10.1371/journal.pone.0124658.
42. Lheureux S, Gourley C, Vergote I, Oza AM. Epithe- 60. Chopra S, Davies EL. Breast cancer. Medicine (United
lial ovarian cancer. Lancet 2019. doi:10.1016/S0140- Kingdom) 2020. doi:10.1016/j.mpmed.2019.11.009.
6736(18)32552-2. 61. American Cancer Society. Breast Cancer Facts & Fig-
43. Xu XJ, Tang YM. Cytokine release syndrome in can- ures 2017-2018. Side Eff Med Cancer Ther Prev Treat
cer immunotherapy with chimeric antigen recep- Second Ed 2018. doi:10.1007/978-3-319-70253-7_2.
tor engineered T cells. Cancer Lett 2014;343:172-8. 62. Shimozuma K. Breast cancer. Japanese J Cancer Chem-
doi:10.1016/j.canlet.2013.10.004. other 2019. doi:10.7748/phc.7.9.17.s18.
44. Dickson BC, Mulligan AM, Zhang H et al. High-lev- 63. Statistic USBC. U.S. Breast Cancer Statistics | Breast-
el JAG1 mRNA and protein predict poor outcome in cancer.org. BreastcancerOrg 2018. doi:10.1007/
breast cancer. Mod Pathol 2007;20:685-93. doi:10.1038/ PL00007975.
modpathol.3800785. 64. Casey SC, Tong L, Li Y et al. MYC regulates the an-
45. Wang H, Peng R, Wang J, Qin Z, Xue L. Circulating titumor immune response through CD47 and PD-L1.
microRNAs as potential cancer biomarkers: The ad- Science 2016. doi:10.1126/science.aac9935.
vantage and disadvantage. Clin Epigenetics 2018. 65. Muller PAJ, Vousden KH. P53 mutations in cancer. Nat
doi:10.1186/s13148-018-0492-1. Cell Biol 2013. doi:10.1038/ncb2641.
46. Vanni I, Alama A, Grossi F, Dal Bello MG, Coco S. Ex- 66. Whibley C, Pharoah PDP, Hollstein M. p53 polymor-
osomes: a new horizon in lung cancer. Drug Discov phisms: Cancer implications. Nat Rev Cancer 2009.
Today 2017. doi:10.1016/j.drudis.2017.03.004. doi:10.1038/nrc2584.
47. Heneghan HM, Miller N, Lowery AJ, Sweeney 67. Eble JN, World Health Organization., International
KJ, Newell J, Kerin MJ. Circulating micrornas as Agency for Research on Cancer., International Acad-
novel minimally invasive biomarkers for breast emy of Pathology. Pathology and genetics of tumours
cancer. Ann Surg 2010;251:499-505. doi:10.1097/ of the urinary system and male genital organs. IARC
SLA.0b013e3181cc939f. Press; 2004.
48. Souza KCB, Evangelista AF, Leal LF et al. Identification 68. Ohno SI, Takanashi M, Sudo K et al. Systemically
of Cell-Free Circulating MicroRNAs for the Detection injected exosomes targeted to EGFR deliver antitu-
of Early Breast Cancer and Molecular Subtyping. J On- mor microrna to breast cancer cells. Mol Ther 2013.
col 2019;2019:1-11. doi:10.1155/2019/8393769. doi:10.1038/mt.2012.180.
49. Yurkovetsky ZR, Linkov FY, Malehorn DE, Lokshin AE. 69. Meyer N, Penn LZ. Reflecting on 25 years with MYC.
Multiple biomarker panels for early detection of ovarian Nat Rev Cancer 2008. doi:10.1038/nrc2231.
cancer. Fut Oncol 2006. doi:10.2217/14796694.2.6.733.
70. Chakrabarti M, Khandkar M, Banik NL, Ray SK. Altera-
50. Resnick KE, Alder H, Hagan JP, Richardson DL, Croce tions in expression of specific microRNAs by combina-
CM, Cohn DE. The detection of differentially expressed tion of 4-HPR and EGCG inhibited growth of human
microRNAs from the serum of ovarian cancer patients malignant neuroblastoma cells. Brain Res 2012;1454:1-
using a novel real-time PCR platform. Gynecol Oncol 13. doi:10.1016/j.brainres.2012.03.017.
2009. doi:10.1016/j.ygyno.2008.08.036.
71. Honeyman JN, La Quaglia MP. Neuroblastoma. Rick-
51. Wang ZH, Xu CJ. Research progress of MicroRNA in ham’s Neonatal Surg, 2018. doi:10.1007/978-1-4471-
early detection of ovarian cancer. Chin Med J (Engl) 4721-3_57.
2015. doi:10.4103/0366-6999.171459.
72. Wilkins MR, Sanchez JC, Gooley AA, Appel RD,
52. Shruthi B, Vinodhkumar P, Selvamani M. Proteomics: Humphery-Smith I, Hochstrasser DF, et al. Progress
A new perspective for cancer. Adv Biomed Res 2016. with proteome projects: Why all proteins expressed
doi:10.4103/2277-9175.180636. by a genome should be identified and how to do it.
53. Read RW. Metastasis. Intraocular Inflamm, 2016. Biotechnol Genet Eng Rev 1996. doi:10.1080/026487
doi:10.1007/978-3-540-75387-2_150. 25.1996.10647923.

JBUON 2021; 26(5): 1733

1734 Molecular cancer indicators

73. American Cancer Society. Cancer Facts and Figures 75. Wang ET, Pisarska MD, Bresee C et al. BRCA1 ger-
2017. Genes Dev 2017;21:2525–38. doi:10.1101/ mline mutations may be associated with reduced
gad.1593107. ovarian reserve. Fertil Steril 2014. doi:10.1016/j.fertns-
74. Stuckey A, Febbraro T, Laprise J, Wilbur JS, Lopes V, tert.2014.08.014.
Robison K. Adherence Patterns to National Comprehen- 76. Bell RA, McDermott H, Fancher TL, Green MJ, Day FC,
sive Cancer Network Guidelines for Referral of Women Wilkes MS. Impact of a Randomized Controlled Edu-
With Breast Cancer to Genetics Professionals. Am J Clin cational Trial to Improve Physician Practice Behaviors
Oncol Cancer Clin Trials 2016;39:363-7. doi:10.1097/ Around Screening for Inherited Breast Cancer. J Gen Intern
COC.0000000000000073. Med 2015;30:334-41. doi:10.1007/s11606-014-3113-5.

JBUON 2021; 26(5): 1734