Molecular genetic changes in gastric carcinoma

INTRODUCTION

Gastric cancer (GC) remains a worldwide burden as one of the leading causes of cancer-related death in both sexes. The late onset of clinical symptoms is the main reason that the disease is often diagnosed at an advanced stage, which limits the available therapeutic approaches in more than 50% of cases.^[1] At present, GC is a most frequent diagnosed neoplasm worldwide. Although extensive studies have been performed to identify genetic pathways and genes involved in the disease development and progression, the prognosis of GC patients remains poor and little improvement of long-term survival has been achieved. Adenocarcinoma is the major histological type of GC, accounting for 90–95% of all gastric malignancies. The incidence is closely related to environmental factors, reflecting a characteristics geographical distribution.^[2]

GC is a result of complex interaction of environment and multiple genes. The evident risk factors of GC include dietary, the Helicobacter pylori infection, the family history, and the genetic factors.^[2,3] Although, there is great progression in the diagnosis and treatment of GC, the survival rate is still stagnant and poor, only about 20% of patients with GC can reach 5-year survival. Thus, a systematic view of the genetic basis of GC is necessary to establish new strategies for prevention and treatment of GC. Genetic factors mainly refer to the susceptible genes of cancer that is involved in multiple genetic and epigenetic alterations of oncogenes, tumor suppressor genes (TSGs), cell cycle regulators, DNA repair genes, and signaling molecules.^[4]

GC is the second cause of cancer mortality^[5] and the fifth most common malignancy in the world, 50% of the cases are from Eastern Asia^[1] where China has the highest incidence.^[6] The 5-year survival rate is still remain disappointing despite improvements in the diagnosis and treatment of GC as they are usually diagnosed at the advanced stage and usually incurable or rarely curable. The environmental risk factors have been established includes H. pylori infection, smoking, consumption of food high in salt, or N-nitroso compounds such as processed or smoked meats and Epstein–Barr virus (EBV) infection. The GC incidence seems to be increasing in younger age groups; however, the cause of phenomenon is still unknown.^[7] Based on the Lauren classification, GC is classified as intestinal and diffuse types, which has different clinicopathological and prognostic features. It not only differs in morphology but also in epidemiology, progression pattern, genetics, and clinical pattern. Recently, it is seen that location of the tumor also matters as there appears to be a difference between proximal and distal non-diffuse GCs because of the distinct or different sets of gene expression level.^[8,9] Surgical treatment is the only therapeutic modality which provides the greatest possibility of cure. Overall, this article aims to cover the molecular and genetic changes, different classification based on several basis genetic and epigenetic changes in GC which are currently understood.

CLASSIFICATION OF GASTRIC CARCINOMA

Early gastric carcinoma isclassified based on gross pattern as follows: Type I for the tumor with protruding growth, Type II with superficial growth, TypeIII with excavating growth, and Type IV for infiltrating growth with lateral spreading. Type II is again divided into IIa (elevated), IIb (flat), and IIIc (depressed) proposed by the Japanese endoscopic society.^[10] Another classification, three gross patterns for superficial neoplastic lesions in gastrointestinal tract, the tumor is classified as: Type 0-I for polypoid growth (subcategorized to 0-IP for pedunculated growth and 0-IS for sessile growth), Type 0-II for non-polypoid growth (subcategorized into Type 0-IIa for slightly elevated growth, Type 0-IIb for flat growth, and Type 0-IIc for slight depressed growth), and Type 0-III for excavated growth.^[11] The difference between intramucosal carcinoma and carcinoma insitu or high grade dysplasia is very important because the stomach intramucosal unlike the colon intramucosal shows metastasis. The useful important histologic features of intramucosal invasion are single tumor cell in lamina propria and significantly fused neoplastic gland of multiple sizes. The early gastric carcinoma prognosis is excellent and shows 5 years survival rate with about 90%.^[12]

Advanced gastric carcinoma shows invasion upto muscularis propria beyond and has showed worse prognosis with 5-year survival rate at about 60% or less.^[13] The gross appearance can be exophytic, ulcerated, infilterative, or combined. Based on Borrmann’s classification, the gross appearance is classified into; Type I for polypoid growth, Type II for fungating growth, Type III for ulcerating growth, and Type IV for diffusely infiltrating growth (linitis plastic in signet ring cell carcinoma), this happens when most of the gastric wall is involved by infiltrating tumor cells. In addition, World Health Organization (WHO) classification endorses other uncommon histologic variants, such as adenosquamous carcinoma, squamous carcinoma, hepatoid carcinoma, carcinoma with lymphoid stroma, choriocarcinoma, parietal cell carcinoma, malignant rhabdoid tumor, mucoepidermoid carcinoma, mixed adeno-neuroendocrine carcinoma, endodermal sinus-tumor, embryonal carcinoma, pure gastric yolk sac tumor and oncocytic adenocarcinoma, and micropapillary carcinoma of stomach is a newly recognized histologic variants characterized by small papillary clusters.^[13]

CLASSIFICATION OF GASTRIC CARCINOMA BASED ON GENETIC PROFILE

According to Laurens classification, gastric adenocarcinoma isdivided into intestinal, diffuse, mixed, and indeterminate subtypes.^[14] They vary not only in morphology but also in epidemiology, progression pattern, genetics, and clinical pictures. Histopathologicaly, intestinal type is characterized by malignant epithelial cells that show cohesiveness and glandular differentiation infiltrating the surrounding tissues.^[15] In contrast, the diffuse subtype is characterized by tumor cells that show poor differentiation and lack of cohesion. Intestinal type of GC is felt to be caused mainly by environmental (exogenous) factors whereas the diffuse type is thought to be due to hereditary and genetic (endogenous) factors. These histologic classifications are not sufficient to reflect the molecular and genetic characteristics of GC or to develop personalized treatment strategies in the era of precision medicine. Several molecular classification systems have been proposed, and distinct molecular subtypes have been identified.^[16-18] In 2014, the cancer genome atlas (TCGA) proposed foursubtypes: (1) EBV-positive (8.8%), (2) microsatellite unstable/instability (MSI, 21.7%), (3) genomically stable (19.7%), and (4) chromosomally unstable/chromosomal instability (CIN, 49.8%).^[16] Most EBV-positive tumors occurred in male patients and in the gastric fundus or body, displaying extreme DNA hypermethylation and amplification of JAK2 and PD-L1/2, with 80% harboring non-silent PIK3CA mutations. All EBV-positive GCs displayed CDKN2A promoter hypermethylation, while lacking the MLH hypermethylation characteristic of the MSI-associated CpG island methylator phenotype (CIMP).^[16,19]

MSI-subtype tumors tends to occur in female patients, diagnosed at advanced stage, and characterized by elevated mutation rates, including mutations of genes encoding targetable oncogenic signaling proteins. The genomically stable sub-type lacked numerous molecular alterations, correlated well with the Lauren diffuse histologic variant, but harbored mutations of RHOA or fusions involving RHO-family guanosine triphosphatase (GTPase)–activating proteins. The active GTP-bound form of RHOA activates signal transducer and activator of transcription-3 (STAT-3) to promote tumorigenesis. Finally, CIN subtype tumors were frequent at the gastroesophageal junction/cardia, correlated well with the Lauren intestinal histologic variant, showed marked aneuploidy, and harbored focal amplifications of receptor tyrosine kinases (RTKs), in addition to recurrent TP53 mutations and RTK-RAS activation.^[16]

In 2015, the Asian Cancer Research Group (ACRG) proposed a new classification system associated with distinct genomic alterations, disease progression, and prognosis across multiple GC cohorts.^[17] On the basis of whole-genome sequencing, gene expression profiling, genome-wide copy number microarrays, and targeted gene sequencing, four molecular subtypes were identified: (1) MSI, (2) microsatellite stable with epithelial-to-mesenchymal transition features (MSS/EMT), (3) MSS/TP53 mutant (MSS/TP53þ), and (4) MSS/TP53 wild-type (MSS/TP53).^[17] MSI tumors were hypermutated, intestinal, usually antral, and diagnosed at clinical Stage I/II. MSI tumors had the best prognosis; their recurrence rate after surgical resection of primary GC was the lowest among all four subtypes (22%). MSS/TP53þ tumors were linked to EBV infection and had the next best prognosis, followed by MSS/TP53- tumors. MSS/EMT tumors occurred at a younger age, mostly diagnosed at clinical Stage III/IV, and were the Lauren diffuse histologic type. The MSS/EMT subtype had the worst prognosis and the highest recurrence rate (63%), with recurrences located mostly in the peritoneal cavity.^[17] To refine molecular classification, regular mutated and hyper mutated GC types were identified in another recent study. The regular mutated type was further sub classified into two subgroups, C1 and C2. The first subgroup, C1, was enriched in mutations of TP53, XIRP2, and APC and was associated with a significantly better prognostic outcome, whereas C2 was over represented by mutations in ARID1A, CDH1, PIK3CA, ERBB2, and RHOA.^[17]

MOLECULAR PROFILE IN GASTRIC CARCINOMA

TCGA has characterized 295 cases of gastric adenocarcinoma, the most comprehensive study to date, using multiple high-throughput technologies, which includes somatic copy number analysis, DNA methylation profiling, messenger RNA, and micro-RNA sequencing, reverse phase protein array, microsatellite instability (MSI), and whole genome sequencing.^[16] After this, four GC subtypes were described they are; (1) Tumor positive for EBV, (2) MSI-high tumors, (3) genomically stable tumors, and (4) tumor with chromosomal instability [Table 1]. Moreover, these GC subtypes showed different epigenetic changes and mutation of several genes like; EBV + tumors show recurrent PIK3CA and ARIDIA mutations, extreme DNA hypermethylation, and high amplification of JAK2, PD-L1, and PD-L2. MSI-H tumors show elevated mutation rates which also include mutation of genes encoding targetable oncogenic signaling proteins.^[17] GS (genomically stable) tumors are enriched by the diffuse histological variants and CDH1 and RHOA mutation or CLDN18-ARCHGAP fusion. CIN tumors observed frequently at the gastroesophageal junction/ cardia along with recurrent TP53 mutation and number of amplifications of RTKs genes.

The ACRG has proposed four molecular subtypes which are as follows; (1) MSH-H, (2) MSS/EMT, (3) MSS/TP53− mutants, and (4) MSS/TP53+ wild type [Table 2].^[17] The recent molecular characterization of GC is evolving. Many molecular classifications being proposed and different molecular subtypes have been identified.^[19] Various RTKs such as human epidermal growth factor (EGF) receptor 2, EGFR1, mesenchymal epithelial transition factor (MET), and fibroblast GF receptor 2 (FGFR2) being reported to be amplified in GC and target therapies have been developed.^[20-23]According to GI-screen as the nationwide cancer genome screening project, the frequently detected mutations are TP53 (47.8%), PIK3CA (9.2%), KRAS (6.0%), SMAD 4 (5.1%), APC (4.1%), TET2 (3.9%), and ERBB2 (3.3%), and copy number variants are ERBB2 (11.3%), CCNEI (11.1%), KRAS (3.7%), FGFR2 (3.3%), ZNF217 (3.3%), MYC (2.7%), CCND1 (2.3%), and CDK6 (2.1%).^[24]

The understanding of the molecular aspects of GC is improved by next generation sequencing (NGS) studies which provides high throughput method to know and identify the genetic alteration in GC systematically. Li-Chang et al. showed mutations of several driver genes by performing NGS that includes; TP53, PIK3CA, CTNNB1, CDH1, SMAD4, and KRAS.^[25] Few of the TSGssuch asAPC, CDH1, CDH4, THBS1, and UCHL1 are found to be inactivated by hypermethylation.^[26-29] It has been shown that 59% of GCs shows mutation in chromatin remodeling genes such asARID1A, PBRM1, and SETD2.^[30] There are new mutated driver genes found, they are (MUC6, CTNN2A, and GLI3) through whole genome sequencing.^[31] It was also found that the genes involved in cell adhesion and chromosome organization showed frequent mutations in gastric adenocarcinoma patients by which it has confirmed that there are 30 candidate driver mutation mutated in primary and lymph node tissues and, on the other hand, seven candidate mutation mutated only in primary tumors. The primary tumors show more mutations than metastatic tumors because metastatic tumors comes only from the subpopulation of primary tumors but surprisingly researchers do not found any metastatic specific mutations and it is expected that these metastatic tumor cell shows the behavior of continuing the mutations of its genome to gain more growth power. There are some locus at chromosome 17q12 which are frequently amplified in GC and they are as follows; PPPIRIB-STARD3-T-CAP-PNMT, PERLD1-ERBB2-MAC14832-GRB7.^[32]

In addition to this, there are two genes which are very important located on 9p21 chromosome which encodes P16 and P15, respectively, they are CDKN2A and CDKN2B showed CN loss (CN=0.8~1.32) as these two genes normally show very crucial function, they inhibit cyclin-dependent kinase (CDK4) and CDK6 and control cellular proliferation by preventing entry into the S phase of the cell cycle and so inactivation of these may leads to uncontrolled growth in cancer.^[33]

GENETIC CHANGES IN GC

Mutation of genes in GC is broadly classified into three categories;

• High frequency drivers which display high rate of recurrence (>5–10%) across multiple GCs

• Low frequency drivers which are recurrently mutated in the 1–10% range but they are still contributing to pathogenesis of disease

• Passenger mutation/bystander which arises as a consequences of underlying mutational processes such as CpG deamination but are not functionally contribute in tumorigenesis.^[34]

Recently, it has been reported the importance of RTK/ RAS/mitogen-activated protein kinase (MAPK) signaling, frequent mutations in the ERBB3 RTK, and the ligand/RTK NRG1/ERBB4 genes in GCs. The recent NGS studies have been highlighted two new GC genes–ARID1A and RHOA. ARID1A is mutated in 10–15% of GCs which is known to encode a components of SWI/SNF chromatin remodellar complex.^[34] Mutation of ARID1A are typically inactivating (frameshifts, and non-sense mutation). Consistently acting as a tumor suppressor genes. The normal function of ARID1A gene is a role in GC cell proliferation through the control of cell cycle regulators CCNE1 and E2F1. The another genes RHOA which has been recently shown that it exhibits recurrent mutations in diffuse type/genome stable GCs.^[34]

The consequences of mutation in both ARID1A and RHOA are different. ARID1A mutation is dispersed throughout the genes, whereas the RHOA mutation is localized to an N-terminal hot-spot region (Ty42, Arg5and Gly17). ARID1A are predicted to modulate downstream Rho-signaling.

Mutation of RHOA suggests after functional studies that they can impart resistance to anoikis (a form of programmed cell death occurring after cellular detachment from a solid substrate). Clinically, RHOA hotspot mutation discovery provides specific inroad for the development of new approaches to target diffuse type GCs (DGCs), traditionally associated with extremely poor prognosis.^[34] Genomic studies also uncovered the long-tail of low frequency driver mutations in GC such as gastric mucin MUC6, BCOR (encoding BCL6 corepressor), FAT 4 (a protocadherin), and RNF43, a Wnt pathway regulator. Although they have low mutation frequency, these genes also contributes to the ability of GC to manifest the cancer hallmarks.^[34]

TSG

TSGs normally show protective role in preventing the malignant transformation of cells by repairing DNA inhibiting cell proliferation and initiating programmed cell death (apoptosis). TSGs are involved in the regulation of a range of cell functions including cell adhesion, cell-cell interaction, cytoplasmic signal transduction, and nuclear transcription.^[35] Over recent decades, there is rapid expansion in the member of TSGs which have been identified in relation to a broad range of inherited and non-inherited human cancers. Greater understanding of the pattern of TSG expression in GC may enables the identification of specific biomarkers which could be used for early diagnosis and development of target treatment. The overexpression of P53 and loss of expression of phosphatase and tensin homolog (PTEN), E-cadherin, SMAD4, MGMT, and CD82 were all significantly associated with poor prognosis in gastric carcinomas.^[36]

1. Tumor protein P53 (TP53): It is the most frequently mutated aberration in ~50% of GC cases. The cellular function of TP53 is a guardian of genomic integrity. The GCs in which TP53 gene is mutated often exhibit high levels of SCNAs involves both broad chromosomal regions and focal gene regions. The mutations in other canonical oncogenes (KRAS, CTNNB1, and PIK3CA) and tumor suppressor genes (SMAD4, APC) have also seen in GCs.^[37] The aberrant expression of TP53 have been found in over half of human cancers including leukemia, breast, colon, and lung carcinoma and alteration in TP53 is the most common event in human cancers. Normally, TP53 plays key role in cell cycle progression preventing G1/S transition after DNA damage occurred and allowing DNA repair/cell apoptosis.^[37] Some studies shown that TP53 abnormalities can occurs in nonneoplastic gastric mucosa with intestinal metaplasia suggesting TP53 mutation as early phenomena in gastric carcinogenesis^[38]

2. PTEN: It is aTSG, discovered in 1997. It functions as a dual-specificity protein and phospholipid phosphatase, and regulates a variety of cellular processes and signal transduction pathways in a complex network system that can control proliferation, apoptosis, migration, adhesion, and genetic stability. The most detailed study of PTEN is focused on its negative regulation on phosphoinositide 3-kinase (PI3K)/Akt signaling pathway, which is important in controlling cell survival, promoting proliferation, and inhibiting apoptosis. It is a lipid phosphatase which phosphorylates phosphotidyl inositol-3,4,5-triphosphate (PIP3) reduces the activation of PI3K and the serin-threonine specific protein kinase Akt.^[39]Hence, loss of expression of PTEN, increased level of PIP3, increase AKT activation, and consequently inhibits apoptosis which ultimately results in large number of tumors. There are multiple phenomena due to which PTEN inactivation occurs such as gene mutation, loss of heterozygosity (LOH), hypermethylation or miRNA-mediated alterations in gene expression, or post-translational modification.^[40] In gastric adenocarcinoma, PTEN protein expression is significantly downregulated compared to normal. On the other hand, PI3K, AKT, MMP-2, MMP-9, and nuclear factor-κB (NF-κBp65) protein is overexpressed in GC^[41]

3. CDH1 (Cadherin 1): Inactivation of E-cadherin found in early stage of gastric carcinogenesis, tumor progression, invasion, and metastasis. CDH1 gene alteration leads to loss of E-cadherin expression which ultimately results in cancer cell proliferation, invasion, and metastasis which is considered as primary carcinogenic event in hereditary diffuse GC.^[42-44] Somatic alterations in CDH1 found in approximately 30% of GC patients. About one-third having structural alterations (7.5% LOH, 1.7% mutation) and rest having epigenetic modifications with 18.4% hypermethylation.^[45] Epigenetic modification is more common than structural in diffuse type while intestinal type has approximately similar rates of structural and epigenetic changes.^[45] Wei et al.^[46] recently found that mRNA and protein expression level of T-cadherin is lower significantly in gastric tumor tissues compared to normal adjacent tissue.^[46] The decreased T-cadherin expression is correlated with larger tumor size, lymph node metastasis, and higher TNM stage.^[46] Multivariate analysis shows that T-cadherin expression is an independent prognostic factor for overall survival^[46]

4. SMAD3 is another tumor suppressor gene whose loss of expression is associated to advanced stage and poorer outcome in gastric carcinoma patients which is more common in intestinal type than in diffuse type gastric carcinoma

5. SMAD4 loss of expression is more common in intestinal type while E-cadherin loss is more common in diffused type.^[36] P53 expression loss is more common in poorly differentiated tumors.^[36] Loss of E-cadherin, SMAD4, CD82, MGMT, and PTEN expression is associated with advanced stage of disease while P16 lossed expression seen in both early and advanced stages.^[36] Loss of TSG expression accumulation in GC tumor expression which has significant correlation between this accumulation and patient survival.^[36] Apart from mutation, genetic instability, activation of oncogene, and aberrant GF expression/receptor activation all together play role in gastric carcinogenesis

6. Retinoblastoma protein 1 (RB1) is involved in negative cell cycle regulation at the G1/S transition. RB1 shows LOH in advanced stage. RB1 is known to be a target of miRNA-106b~25 which is involved in promotion of cell proliferation, EMT, cell cycle progression, and exerts an apoptotic effect^[47]

7. Pre-myelocytic leukemia is a TSG controls apoptosis and cell proliferation through P73 and yes associated protein-1 which are found downregulated in GC^[47]

8. Fbxw7 is a TSG that is responsible for the degradation of several proto-oncogenes and its functional inactivation can dysregulate the cell division process, and potentially lead to tumorigenesis. The inactivation of Fbxw7 in the progress of human cancer including mutation, deletion, and hypermethylation, of which the Fbxw7 mutation is most common.^[48] The Fbxw7 mutation rate in GC tissues was from 3.7% to 6% and did not differ in early or advanced GC, which might play a role in the prognosis of GC.^[49] Fbxw7 mRNA expression level in cancerous tissues was lower than that in noncancerous tissues. The low Fbxw7 expression had remarkable poorer prognosis than those with high Fbxw7 expression.^[50] Fbxw7 expression was associated with the progressive tumor size, lymph node metastasis, peritoneal dissemination, venous invasion, and clinical stage. Fbxw7 induced tumor apoptosis and growth arrest and inhibited the EMT in part by downregulating the RhoA signaling pathway in GC.^[51]

RAS

The first human oncogene identified which is involved in 20% of all human malignancies. These are the oncogenes which are not present in normal cells rather they are activated in tumor cells as a consequences of enormous mutations occurs during tumor development. RAS gene encodes guanine-nucleotide binding proteins which perform various functions in mitogenic signal transduction and protein activity of RAS is controlled by GTP or GDP binding (active – GTP bound and inactive-GDP bound) states.

LOH (Loss of Heterozygosity)

This describes a genetic phenomenon often seen with tumor suppressor genes in cancer. Since the human karyotype is diploid, mutation of one allele of a tumor suppressor gene is not sufficient to cause cancer. In heterozygous individuals, the wildtype allele will provide for a functional phenotype. However, when a“second hit” occurs, for example, through mis-segregation of chromosomes, this individual (or cell) may lose its “heterozygosity,” leading to a full cancerous phenotype. Recently, Karaman et al.^[107] found significant correlation between prevalence of 17p (TP53) LOH in gastric precancerous lesions, indicating that loss of TP53 could be an early event in gastric carcinogenesis.^[107]

Research in recent years demonstrated that although PTEN mutations in GC are rare, LOH is more frequent. Byun et al. (2003) found decreased expression of PTEN and upto 47% LOH in 33% (5/15) GC cell lines and 36% (22/55) GC tissue samples.^[108] LOH rate was significantly higher in advanced than in early GC (63–18%); it was also significantly higher in poorly differentiated than highly and moderately differentiated GC (69–29%). This suggests that complete functional inactivation of PTEN is not necessary to cause gastric carcinogenesis, loss of one allele is sufficient.^[108] LOH occurs in different stages of GC, genetic mutations could be detected only in advanced GC. Both gene mutations and LOH can promote tumor invasion and metastasis. Some studies have shown that the LOH status of PTEN can impact sensitivity to chemotherapy and hence, the prognosis in patients with GC. It may also be used as an independent prognostic tool. There are several putative TSGs, the hypermethylated in cancer 1 (HIC1) gene and TOB1 gene. The HIC1 gene is an interesting candidate TSG located in the overlapping LOH subregion of LOH in chromosome 17p13.3. It encodes a zinc-finger transcription regulator that contains an NH₂-terminal BTB-POZ domain characteristic of a family of repressors. The gene is ubiquitously expressed in normal tissues, but a decrease or loss of expression was found in several types of tumors excluding GC.^[109] Indeed, there is accumulating evidence indicating that a change in HIC1 expression through epigenetic mechanisms has an important role in tumor progression. The patterns of monoallelic methylation associated with a partial loss of HIC1 expression and biallelic methylation, together with a marked loss in HIC1 expression, were seen in these GCs. The concomitant loss or decreased expression and promoter methylation of HIC1 indicated that HIC1 may be silenced by aberrant DNA methylation, and could be a target for inactivation by epigenetic events in gastric tumorigenesis.^[110] Another notable candidate gene in the overlapping LOH subregion in chromosome 17q21.33 (SR₃) is TOB1 (transducer of ERBB-2, 1). TOB1 is a member of the TOB/BTG antiproliferation protein family. Overexpression of the TOB family proteins results in cell growth retardation,^[109] while TOB LOH and decreased TOB expression were observed in lung cancer tissues.^[110]

Genetic analyses of LOH helped to identify the chromosomal location of many tumor suppressor genes. LOH analyses have identified several arms and regions of chromosomes that contain or potentially harbor tumor suppressor genes important in gastric tumorigenesis. Chromosome 3p was the most frequently lost locus in LOH analysis of over 100 archived stomach cancers.^[111] Moreover, three distinct regions of chromosome 4q were found to be frequently lost in gastroesophageal junctional adenocarcinomas.^[111] LOH is also a marker of chromosomal instability and might indicate a second inactivation hit of cancer suppressor genes. Several LOH studies demonstrated that the extent of chromosomal loss appeared to be of prognostic significance. LOH has been shown to relate to cancer progression, where a transition from LOH-L to LOH-H is thought to reflect an increase in chromosomal instability during tumor advancement.

The highest LOH frequencies have been identified at 1p, 2q, 3p, 4p, 5q, 6p, 7p, 7q, 8p, 9p, 11q, 12q, 13q, 14q, 17p, 18q, 21q, and 22q chromosome regions in gastric carcinomas.^[111] The main consequence of LOH is loss of genes, such as tumor suppressors, cell cycle regulators, DNA repair genes, and other genes implicated in the maintenance of cell cycle, and/or integrity of DNA. LOH studies have revealed that allelic imbalance of chromosome 17 loci occurs frequently in sporadic GC. Most of the LOH studies performed on chromosome 17 to date have used only a limited number of markers, and a detailed deletion map of overlapping regions on chromosome 17 has not yet been made in sporadic GC.

HYPERMETHYLATION

DNA methylation is a reversible chemical modification of the cytosine in the CpG islands of promoter sequences, catalyzed by a family of DNA methyltransferases. DNA methylation does not change the genetic information but it just alters the readability of the DNA and results in the inactivation of gene by subsequent transcript repression.^[125] In general, CpG island methylation is associated with gene silencing. The methylated CpG Island also recruits histone deacetylases (HDAC) and other factors involved in transcriptional silencing.^[122] Inactivation of TSGs through hypermethylation of CpG islands within promoter regions is a major event in carcinogenesis.^[122] It includes various genes involved in different cellular process such as cell cycle regulation (p16^NK4a, CDKN2b/p15^INK4b, and p14ARF) which is hypermethylated in human cell lines and primary tumors, DNA repair genes (hMLH 1 and MGMT) were observed aberrantly methylated at their promoter regions, cell-cell/ cell matrix adhesion genes (E-cadherin, H-cadherin and adenomatous polyposis coli [APC]), apoptosis (death-associated protein kinase [DAPK], TMS1, and Caspase-8), and angiogenesis (THBS-1 and p73).^[126] Silencing of p16^INK4a by promoter hypermethylation has also been reported in gastric carcinoma. CDKN2A hypermethylation may contribute to the malignant transformation of gastric precursor lesions. Hypermethylation of DAPK was observed in intestinal, diffuse, and mixed type of GC and correlated with the presence of LN metastasis, advanced stage, and poor survival.^[127] The epigenetic silencing of XAF1 gene by aberrant promoter methylation is reported in GC.^[128] Caspase-1 (IL-1 beta-converting enzyme), a member of the cysteine protease family, shows loss of expression in 19.3% cases of gastric carcinoma^[129] and the expression is reversed on 5-aza 2’deoxycytidine and/or trichostatin treatments in GC cell line. The expression of TSPYL5 mRNA was frequently downregulated due to hypermethylation of its CpG Island.^[129] Recently, Sepulveda et al. (2016)^[130] identified 13 genes (BRINP1, CDH11, CHFR, EPHA5, EPHA7, FGF2, FLI1, GALR1, HS3ST2, PDGFRA, SEZ6L, SGCE, and SNRPN) which are hypermethylated in gastric carcinoma as compared with normal mucosa, which in turn may be useful in developing diagnostic and prognostic tool for this lethal malignancy.

Hypomethylation of specific genes also contribute to gastric carcinogenesis. Global hypomethylation of the genome was initially thought to be an exclusive event in cancer development.^[131] The loss of methylation in cancer is mainly due to hypomethylation of repetitive DNA sequences. During the development of neoplasm, the degree of hypomethylation of genomic DNA increases as the lesion progress from a benign disease to metastatic.^[122,132] Three mechanisms of DNA hypomethylation have been proposed in the development of cancers. First, it increases the genomic instability, second by reactivation of transposable elements, and third by loss of imprinting. Demethylation of DNA can favour mitotic recombination, leading to deletions, translocations, and chromosomal instability.^[122] Demethylation of MAGE, synuclein-γ (SNCG), and cyclin D2 has been described in gastric carcinoma.^[133] SNCG demethylation is common in cases with LN metastasis.^[130] Hypomethylation of cyclin D2 promoter was found in 71% cases of gastric carcinoma. This event is more common in Stage-III and IV tumor than in Stage-I and II tumor.^[134]

In parallel to global hypomethylation, hypermethylation of CpG Island also has silencing effect on miRNAs. MicroRNAs are short, 18–22 nucleotide, noncoding RNAs that regulate many cellular functions including cell proliferation, apoptosis, and differentiation by silencing specific target genes through translational repression or mRNA degradation.^[127,126]

HISTONE MODIFICATION

In normal cell, a precise balance maintains nucleosomal DNA in either an active/acetylated or an inactive/deacetylated form. This adequate balance is controlled by acetylating enzymes (histone acetyltransferases) and deacetylating enzymes (HDACs). The other modification includes methylation of arginine and lysine residues of histones. This methylation is catalyzed by histone methyltransferase and the process is involved in the regulation of a wide range of gene activities and chromatin structures. In general, lysine methylation at H3K9, H3K27, and H4K20 is associated with gene silencing whereas methylation at H3K4, H3K36, and H3K79 isassociated with gene activation.^[124]

Increasing evidence suggests that epigenetic changes play a key role in cancer development including GC. CpG island methylation phenotype as high CIMP (CIMP-H) play an important role in gastric carcinoma progression.

HEREDITARY GC

While the great majority of GCs is sporadic, familial aggregation occurs in about 10%of the cases and out of these only 1–3% clearly constitutes hereditary form. Hereditary GC includes syndromes such as hereditary diffuse GC, gastric adenocarcinoma and proximal polyposis of the stomach (GAPPS), and familial intestinal GC (FIGC). GC has also been identified as part of other hereditary cancer syndromes such as hereditary non-polyposis colorectal cancer, Li-Fraumeni syndrome, familial adenomatous polyposis, and Peutz-jeghers syndrome.^[135]

HDGC is one of the best genetically characterized forms of hereditary GC. Heterozygous germline CDH1 (E-cadherin) mutations, including frameshifts, splice site, nonsense, and missense mutations as well as large rearrangements, were until recently the only unknown causative alterations of HDGC representing upto 40% of patients belonging to families that fulfill the clinical criteria for HDGC.^[136] Germline variants of genes also cause GCs.^[137] Hereditary diffuse GC is a well-known familial GC which is linked to CDH1 (E-cadherin) gene variants.^[137] The physical position of the germline mutation of CDH1 gene vary according to ethnic and geographical differences.^[137] There are 100 variants of the CTNNA1 (α-catenin), a CDH1 binding protein also cause HDGC.^[138]

HDGC is an autosomal dominant cancer susceptibility syndrome characterized by DGC which is principally caused by inactivating germline mutations in CDH1 gene.^[139] CDH1 gene encodes E-cadherin has major function in cell-cell adhesion, maintenance of epithelial architecture, cell polarity, and regulation of intracellular signaling pathways. The pathogenic mutation in CDH1 shows 70% risk of developing DGC by age of 80 years by birth.^[140] Histopathology of advanced HDGC is comparable to sporadic DGC, though the presence of typical precursor lesions, insitu or pagetoid signet-ring cells, isspecific for CDH1- mutation related HDGC. Early HDGC shows indolent (showing no real interest/efforts) phenotype, while advanced HDGC display an aggressive phenotype with a mixture of pleomorphic cells, increased proliferation and aberrant P53 expression.^[141] There are 10–40% of families who fulfills current genetic testing criteria have CDH1 germline mutations.^[114] Although it is not known exactly how many HDGC families with germline CDH1 mutations have been diagnosed worldwide to date, over 155 different CDH1 germline mutations have been reported in HDGC families.^[142] These mutations span the length of CDH1 and no major hotspots have been identified, although some including c.1137G>A and c.792C>T have been observed in several unrelated families.^[140,142]

CDH1 mutation positive patients who postponed a prophylactic gastrectomy, but underwent regular surveillance endoscopy, had a ten-fold increased risk of DGC compared with individuals at risk of HDGC without a mutation.^[143] CDH1 negative families, fulfilling HDGC criteria, often present monoallelic expression of CDH1 in the germline, which mimics functionally the presence of a CDH1 truncating mutation.^[145]The underlying genetic mechanism causing this expression imbalance is still elusive, this finding hints toward CDH1 involvement in a larger fraction of HDGC families. The complete inactivation of the CDH1 gene required for tumor initiation occurs mainly through promoter methylation in primary cancer and loss-of-heterozygosity in lymph node metastases.^[145,146] The ongoing search for novel HDGC predisposing germline aberrations identifies new variants in candidate genes using multiplexed panel and whole exome sequencing.^{[140,147-149]} Germline CTNNA1 mutations have been found in families with HDGC.^[149,150] CTNNA1, encoding α-catenin, is involved in cell adhesion, and forms a complex with β-catenin to bind the cytoplasmic domain of E-cadherin to the cytoskeleton.^[150] New hereditary GC genes are the most promising genes which seems to be the DNA double-strand break repair genes PALB2, BRCA2, RAD51C, and ATM,^{[140,147-151]} which earlier reported in breast cancer families. PALB2 and RAD51C are critical in homologous recombination which is a major DNA repair pathway. While designing gene panels to identify HDGC, these new candidate genes can be considered to broaden the understanding of genetic causes of this disease.

Syndromes with fundic gland polyps and increased GC risk include GAPPS, an autosomal-dominant heritable form of GC, which is caused by point mutations in the APC gene promoter 1B.^[152,153] Affected patients shows specific clinicopathologic phenotype of fundic gland with about 10–100s of polyps involving oxyntic mucosa, occasional hyperplastic, and adenomatous polyps, with an increased risk for intestinal-type (WHO tubular type) or mixed-type gastric adenocarcinoma.^[152] The gastric antrum and pylorus are typically spared. In this syndrome other histological features involves hyperproliferative aberrant pits, characterized by disorganized proliferations of oxyntic glands high up in the mucosa with an increased proliferative activity.^[154] The dysplasia in GAPPS is of gastric foveolar phenotype with positive staining for MUC5AC and MUC6.^[154]

FIGC intestinal-type gastric carcinoma patients have also shown familiar clustering. The diagnosis in young patients is strongly associated with a family history of GC in first degree relatives.^[155] Familial aggregation of H. pylori infection seems to be related to incidence of intestinal-type GC.^[155] The diagnosis should be considered when there is a history of intestinal-type GC in families without polyposis. A single family with clustering of intestinal-type GC and heterozygous, mutations in the immunity gene IL12RB1 has been reported, but further research is required to determine whether such mutations increase GC risk.^[156]

FIGC follows an autosomal dominant inheritance pattern^[158] and diagnostic criteria^[146] are stratified on the basis of the population GC incidence. In low-incidence countries, diagnosis requires that a patient has at least two first-degree or second-degree relatives with intestinal GC, one by 50 years of age, or has three or more first-degree or second-degree relatives diagnosed with GC at any age.^[158,159] In high-incidence countries, diagnosis requires three criteria: (1) At least three relatives with intestinal GCs with one of them being a first-degree relative of the other two, (2) another criteria are occurrence of GC in at least two generations, and (3) and diagnosis of GC before 50 years of age in at least one patient.^[95,159]

CONCLUSION

GC is a collection of different molecular entities and its landscape is enormously complex, but recent efforts and knowledge enabled for the development of more modern and solid therapeutic approaches. Several important advances have been achieved in GC genetics. The improvement in our understanding of the molecular genetics of GC has greatly hastened over the last decades, enabling us to redefine the disease at the molecular level. These findings may lead to the identification of high-risk groups that can be targeted for early interventions which may improve our understanding of tumorigenesis and will ultimately lead to improved outcomes for this common malignancy. Molecular classifications, especially TGCA and ACRG, opened the doors wide on the complete comprehension of the complex genetic landscape of GC. Recent genomic and epigenomic profiling studies provides an improved molecular understanding of GC. In the above article, the characterization and classification of GC at molecular and genetic level would further support that this disease is highly heterogeneous. Researchers and clinicians should take advantage of the information provided by these studies to both design and test potential screening markers and new targeted therapeutic approaches.