Potential prognostic, diagnostic and therapeutic markers for human gastric cancer

Abstract

The high incidence of gastric cancer (GC) and its consequent mortality rate severely threaten human health. GC is frequently not diagnosed until a relatively advanced stage. Surgery is the only potentially curative treatment. Thus, early screening and diagnosis are critical for improving prognoses in patients with GC. Gastroscopy with biopsy is an appropriate method capable of aiding the diagnosis of specific early GC tumor types; however, the stress caused by this method together with it being excessively expensive makes it difficult to use it as a routine method for screening for GC on a population basis. The currently used tumor marker assays for detecting GC are simple and rapid, but their use is limited by their low sensitivity and specificity. In recent years, several markers have been identified and tested for their clinical relevance in the management of GC. Here, we review the serum-based tumor markers for GC and their clinical significance, focusing on discoveries from microarray/proteomics research. We also review tissue-based GC tumor markers and their clinical application, focusing on discoveries from immunohistochemical research. This review provides a brief description of various tumor markers for the purposes of diagnosis, prognosis and therapeutics, and we include markers already in clinical practice and various forthcoming biomarkers.

Core tip: Serum-based gastric cancer tumor markers and their clinical significance or application are discussed. Serum-based carcinoembryonic antigen, carbohydrate antigen 19-9 and carbohydrate antigen 72-4 and tissue-based human epidermal growth factor receptor 2/Neu are potential tumor markers for various types of gastrointestinal cancer. This review provides a brief description of various tumor markers for the purposes of diagnosis, prognosis and therapeutics, and we include markers already in clinical practice and various forthcoming biomarkers. Hopefully, based on the markers, we will generate accurate diagnoses, prognoses and select the most appropriate therapy.

INTRODUCTION

Gastric cancer (GC) is the second most common cancer worldwide and the sixth leading cause of tumor-associated death in Taiwan. However, the pathogenic mechanisms underlying GC tumorigenesis remain unknown[1]. Surgery continues to be the only curative treatment for GC. In a recent study, more than 30% of surgical patients who presented with GC were too advanced for curative resections. To improve the low survival outcome and assist in the earlier diagnosis of specific tumor types, new prognostic and/or therapeutic tumor markers are required[2]. Indeed, the identification of novel tumor markers presents a new approach for the early screening of populations and for GC therapy.

In this review, we focus primarily on gastric adenocarcinomas. Based on Lauren’s classification, “intestinal-differentiated” (I-GC) and “diffuse-undifferentiated” (D-GC) are two histomorphologic GC subtypes[3,4]. Although these subtypes show related histomorphologic lesions, the tumor cells may contrast in their aggressiveness or response to chemotherapy[5]. The molecular actions involved in the progression of GC are multipart and involve multiple genes and steps that operate sequentially or in concert[4]. Several risk factors for GC, including Helicobacter pylori (H. pylori) infection, genetic alterations, and chromosomal instability, have been reported[6-8]. The identification of various tumor markers has also added to our basic knowledge of molecular and cellular mechanisms of GC tumorigenesis and progression[8]. The majority of tumor markers are effective prognostic tools that are used to identify groups of patients at risk of relapse or metastasis or to monitor cancer survivors following treatment[9]. However, suitable tumor markers to expose the molecular mechanisms of GC or monitor disease development are critical.

Cancer may increase tumor markers in blood or body tissues. There are numerous and varying tumor markers for specific disease processes, which are used to aid in the diagnosis of cancer. An increase in a particular biomarker may indicate cancer. However, there may also be other underlying causes. Tumor markers may be produced directly by cancer cells or by non-cancer cells responding to the tumor. In summary, the diagnostic assay should include gastroscopy with biopsy. Thus, tumor markers are not optimal for screening for GC on a population basis, and as diagnostic assays, they show poor sensitivity and specificity. Assays based on multiple tumor markers will provide more exact results[10]. Thus far, assays based on multiple tumor markers are helpful only as prognostic indicators in patients with GC to monitor cancer survivors following treatment[11-15]. Recently, many studies have been performed using microarrays/proteomics, and they have reported new markers with potential clinical significance in patients with GC. In this paper, we review the progress in serum-based tumor markers associated with GC, focusing on discoveries from microarray/proteomics research[16-20]. The most widely investigated serum-based tumor markers for GC are listed in Table 1.

Table 1.

Serum-based tumor markers in gastric cancer

Biological category	Biomarker (Gene name)	Cutoff level /Expression levels in human GC serum	Clinical significance	Associated tumor types	Ref.
Oncofetal antigen	CEA	5 μg/L	Prognosis	Gastrointestinal cancer	[24,25]
			GC progression monitoring	Cervix cancer
				Lung cancer
				Ovary cancer
				Breast cancer
				Urinary tract cancer
Glycoprotein antigen	CA19.9	37 kU/L	Prognosis	Mainly pancreatic cancer	[26]
			GC progression	but also colorectal cancer
			monitoring	other types of gastrointestinal cancer
	CA72.4	6 kU/L	Prognosis	Ovary cancer	[27]
			GC progression	Breast cancer
			monitoring	Colon cancer
				Pancreatic cancer
Cytokeratins	Cyfra21.1	Variable	Prognosis	Many types of carcinoma	[29]
				Some types of sarcoma
	TPA	Variable	Prognosis	Carcinoma	[30]
	TPS	Variable	Prognosis	Adult epithelial tumours	[30]
Hormone	Β subunit of hCG	4 μg/L	Prognosis	Seminoma	[31]
				Choriocarcinoma
				Germ cell tumors
				Choriocarcinoma
				Islet cell tumor
				Testicular cancer
Cell proliferation	Reg IV	Overexpression	Prognosis	Gastrointestinal malignancies	[33,34]
	Granulin	Overexpression	Prognosis	Gastrointestinal malignancies	[35,36]
Signaling protein	14-3-3 protein	Downexpression	Prognosis	Epithelial cancer	[40-42]
				Lung cancer
				Breast cancer
				Bladder cancer
				Liver cancer
				Oral cancer
				Head-and-neck squamous cell cancer
	DKK-1	Overexpression	Prognosis	Prostate cancer	[43-49]
				Breast cancer
				Colon cancer
				Esophageal cancer
				Lung cancer
				Multiple myeloma (MM)
Inflammatory	SAA	Overexpression	Prognosis	Lung cancer	[50,52,53]
				Kidney cancer
				Colon cancer
				Prostate cancer
				Liver cancer
Cell apoptosis	MG7-Ag	Overexpression	Prognosis	Gastrointestinal cancer	[55,56]
Enzyme	PG	Overexpression	Prognosis	Gastrointestinal cancer	[57]

CEA: Carcinoembryonic antigen; CA19-9: Carbohydrate antigen 19-9; CA72-4: Carbohydrate antigen 72-4; Cyfra21.1: Cytokeratin subunit 19 fragment; TPA: Tissue polypeptide antigen; TPS: Tissue polypeptide-specific antigen; β-hCG: β-subunit of human chorionic gonadotropin; Reg IV: Regenerating gene IV; Dkk-1: Dickkopf-1; SAA: Serum Amyloid A; MG7-Ag: Gastric cancer associated antigen; PG: Pepsinogens.

CARCINOEMBRYONIC ANTIGEN

Carcinoembryonic antigen (CEA) is a cell-surface-anchored protein involved in cell-cell adhesions. CEA serves as a functional receptor for colon cancer E-selectin and L-selectin ligands, which may be critical for the metastatic spreading of colon cancer cells[21,22]. Serum from individuals with colon cancer often shows higher levels of CEA than that from healthy controls. CEA from serum is primarily used as a biomarker to monitor colon cancer treatment, to identify recurrences following surgical resection and to stage or localize cancer spread[23].

CEA levels may also be raised in other cancer types and in some non-neoplastic conditions. Several factors may influence increases in CEA. It is associated with tumor grade, lymph node metastasis, distant metastasis and tumor stage, suggesting that as the tumor progresses, CEA levels increase[24].

Anti-CEA antibodies are also frequently used in IHC to identify cells expressing CEA in tissue samples. In adults, CEA is expressed only in cancer cells, primarily adenocarcinomas. It can therefore be used to distinguish between this and other similar types of cancers. Because even anti-CEA antibodies tend to show some degree of cross-reactivity, false positive results are observed, and this assay is typically used in combination with other tests[25].

CARBOHYDRATE ANTIGEN 19-9

In 1981, carbohydrate antigen 19-9 (CA19-9) (also called the Lewis antigen) was first discovered in the serum of individuals with colon or pancreatic cancer. However, CA19-9 is used primarily in the management of pancreatic cancer. CA19-9 may be increased in many types of gastrointestinal cancer[26].

CARBOHYDRATE ANTIGEN 72-4

Carbohydrate antigen 72.4 (CA72.4) is a mucin-like glycoprotein found on the surface of many cancer cells[27]. CA72.4-Ab assay shows good specificity for GC and is used to identify GC relapses and for follow-up after treatment[28].

CYTOKERATIN SUBUNIT 19 FRAGMENT, TISSUE POLYPEPTIDE ANTIGENAND TISSUE POLYPEPTIDE-SPECIFIC ANTIGEN

Cytokeratin subunit 19 fragment (Cyfra21.1) is potentially useful for monitoring lung carcinoma. It is a member of the keratin family. The keratins are intermediate filament proteins responsible for the structural integrity of epithelial cells and are grouped into cytokeratins and hair keratins. Cyfra21.1 has been suggested to play a role in lung, colon, stomach, pancreas, breast, and prostate cancers. Cyfra21.1 is the most sensitive tumor marker for non-squamous cell lung cancer. Because Cyfra21.1 detects only fragments of cytokeratin 19, the test shows higher specificity than tissue polypeptide antigen (TPA) and is an independent prognostic factor[29].

The TPA tumor marker comprises a molecular complex of cytokeratins 8, 18, and 19. It is used in the diagnosis and staging of bronchogenic cancer. TPA assays represent first-generation cytokeratin tumor marker tests.

Although tissue polypeptide-specific antigen has been reported to be a potentially useful serum marker in adult epithelial tumors, few reports have been published on childhood malignancies. Currently, there is no widely used marker for Wilms’ tumor[30].

β-SUBUNIT OF HUMAN CHORIONIC GONADOTROPIN

β-subunit of human chorionic gonadotropin (β-hCG) is a hormone formed by the syncytiotrophoblast, a component of the fertilized egg, following conception. Following grafting, the syncytiotrophoblast gives rise to the placenta. Some tumors produce β-hCG; thus, elevated levels in individuals who are not pregnant may indicate cancer. However, whether β-hCG production promotes carcinogenesis remains unknown.

β-hCG may be used as a tumor marker as its β subunit is secreted by some tumors. Based on this reason, a positive result in males indicates testicular cancer. Combined with α-fetoprotein, β-hCG is a good tumor marker for monitoring of tumors of germ cells[31].

REGENERATING GENE IV

Regenerating gene IV (Reg IV) belongs to the regenerating family of secreted C-lectin proteins[32]. Reg proteins are expressed normally in the gastrointestinal tract and are induced in inflammatory bowel disease and in some gastrointestinal malignancies. They show multiple functions including enhancing tissue regeneration, proliferation and are anti-apoptotic[33]. Reg IV is expressed at low levels in a subgroup of primary tumors and is moderately or highly expressed in a majority of hormone refractory and metastatic tumors[34].

GRANULIN

Granulins are a family of secreted, glycosylated peptides that are cleaved from a single precursor. Both the peptides and the intact granulins regulate cell growth. However, different members of the granulin family may act as inhibitors, activators, or show dual roles in cell growth. The granulin family is important for normal development, wound healing, and in tumorigenesis[35]. In addition, the human liver contributes to the development of liver cancer by secreting a granulin-like growth hormone[36].

14-3-3 PROTEINS

14-3-3 proteins, which belong to a family of regulatory proteins, are able to bind a multitude of signaling proteins of various functions, such as kinases, phosphatases, and transmembrane receptors. More than 200 signaling proteins have been reported as being ligands for 14-3-3 proteins[37,38].

Alterations in the expression of several 14-3-3 proteins have been associated with some human cancers. The down-regulation of 14-3-3r has been reported in a multitude of human epithelial cancers. Recently, Urano et al[39] showed that breast cancer cells reduced 14-3-3r protein levels by upregulating the protein Efp, a ring-finger-dependent ubiquitin ligase (E3), which targets 14-3-3r for ubiquitin-mediated degradation by the proteosome. MCF7 cells treated with antisense Efp constructs showed increased levels of 14-3-3r, and athymic mice transplanted with these cells showed decreased tumors. Similar trends of cancer-associated downregulation of 14-3-3r have been reported in human lung cancer, vulva squamous neoplasias, bladder cancer, liver cancer, oral cancer, and head-and-neck squamous cell cancer, suggesting a common role for 14-3-3r as a tumor suppressor[40-42].

DICKKOPF-1

Dickkopf-1 (DKK1) inhibits Wnt/β-catenin signaling, which is essential for embryonic head development. DKK1 regulates Wnt signaling by binding to the Wnt coreceptor lipoprotein-related protein-5 (LRP5). Overexpression or down-regulation of DKK1 gene expression levels has been observed at various stages of tumorigenesis in multiple cancers, such as breast, colorectal, prostate, esophageal, lung cancers, and multiple myeloma (MM). Thus, DKK1 play dual roles as an oncogene or tumor suppressor. Serum from individuals with breast cancer and bone metastases showed significantly higher DKK1 levels compared with healthy individuals. In addition, human primary lung and esophageal cancers, lung cancer cell lines, and esophageal cancer cell lines show increased DKK1 expression. However, DKK1 protein expression was also down-regulated in colon cancer and melanoma[43-49]. Qian et al[44] reported that DKK1 protein expression was detected in the majority of MM cell lines and clinical MM samples.

SERUM AMYLOID A PROTEIN

Serum amyloid A (SAA) proteins combined with the high-density lipoprotein (HDL) complex belong to a family of apolipoproteins that are secreted during the acute phase of inflammation[50]. There are three isoforms of SAA. SAA (SAA1 and SAA2) levels are consistently increased in the liver during acute inflammation, and SAA3 is induced in various distinct tissues[50,51].

SAA are not specific for any type of cancer and are expected to be elevated in other malignant diseases and in inflammatory diseases[52,53]. Cho et al observed that SAA had previously been reported to increase in several different cancers including kidney, colon, prostate cancers, leukemias and lymphomas. However, these proteins are released into the blood from the liver. These findings suggest that these proteins may not be of much clinical significance in our future efforts to more effectively diagnose and monitor cancer[50].

GASTRIC CANCER ASSOCIATED ANTIGEN (MG7-AG)

Akashi et al[54] produced a MG7-Ag monoclonal antibody to use for GC screening. Using the immuno-polymerase chain reaction (Immuno-PCR) assay, the authors observed that MG7-Ag had the highest sensitivity and specificity for detecting GC. They tested serum MG7-Ag levels and found that the rate of MG7-Ag positivity was 82.8% for GC in 198 GC patients and 44.4% for colonic cancer. However, the Immuno-PCR assay is too complicated and expensive for screening the high-risk population in Linqu county. Therefore, they used an ELISA to detect MG7-Ag expression in serum from patients with GC or other carcinomas. They also used IHC to test MG7-Ag expression in GC tissues[55,56].

PEPSINOGENS

There are three genes that encode human pepsinogen A. A fourth human gene encodes gastricsin, also known as pepsinogen C, a digestive enzyme. Gastricsin is an aspartic proteinase that belongs to the A1 peptidase family and is produced in the stomach; it constitutes a major component of the gastric mucosa. Gastricsin is secreted into the serum and is synthesized as an inactive zymogen that includes a highly basic prosegment. Gastricsin is converted into its active mature form at low pH by serial cleavage of the prosegment, performed by the enzyme. Polymorphisms in gastricsin are associated with susceptibility to GC. Serum levels of the enzyme have been identified as tumor markers for specific GCs and H. pylori-associated gastritis[57].

Tumor markers may also be detected using IHC. Although IHC remains the most reliable and cheapest method, several efforts have been made to identify and test novel tumor markers to achieve the prognosis goals. In recent years, many proteins have been identified and validated for their clinical significance in the management of GC. The most widely investigated tissue-based tumor markers for GC and the most correlated clinical parameters are listed in Table 2.

Table 2.

Tissue-based tumor markers in gastric cancer

Biological category	Biomarker (Gene name)	Parameters	Clinical significance	Associated tumor types	Ref.
Receptor	HER2/neu	Lauren histotype	Prognosis	Ovarian cancer	[58-63]
		Lymph node metastasis	Therapeutic response	Stomach cancer
				Uterine cancer
				Endometrial cancer
				Breast cancer
				Esophageal cancer
	VEGF	Lauren histotype	Prognosis	Breast cancer.	[64-68]
		Tumor progression	Therapeutic response	Colon cancer
				Gastric cancer
				Liver cancer
				Lung cancer
Ion channel	hERG1	Lauren histotype	Prognosis	Endometrial cancer	[69]
		Grading		Gastric cancer
		Stage		Colorectal cancer
Transcription factor	KLF5/IKLF/BTEB2	Grading	Prognosis	Bladder cancer	[71,72]
		Stage		Prostate cancer
		Lymph node status		Breast cancer
				Colon cancer
	SATB1	Lymph node metastasis	Prognosis	Breast cancer	[75-78]
		Distant metastasis		Gastric cancer
		Stage		Colon cancer
				Bladder cancer
	c-myc2	Lymph node metastasis	Prognosis	Cervix cancer	[80]
				Colon cancer
				Breast cancer
				Lung cancer
				Stomach cancer
Enzyme	CA IX	Lymph node metastasis	Prognosis	Kidney cancer	[81-83]
				Colon cancer
				Cervical cancer
	MMP-2	Stage	Prognosis	Ovarian cancer	[79,80,84,85]
		Survival		Colon cancer
				Breast cancer
				Bladder cancer
				Gastric cancer
	HDAC	Lymph node metastasis	Prognosis	Renal cancer	[86-92]
		Survival		Colorectal cancer
				Gastric cancer
				Breast cancer
				Pancreatic cancer
				Liver cancer
				Lung cancer
				Prostate cancer
				Hodgkin's lymphoma
	COX-2	Lymph node metastasis	Prognosis	Breast cancer	[93,94]
		Survival		Intestinal tract cancer
	SLPI	Depth of invasion	Prognosis	Ovan cancer	[96,97]
		Serosal invasion		Cervical cancer
		Lymph node metastasis		Gastric cancer
		Stage
	GLO1	Depth of invasion	Prognosis	Colon cancer	[99-103]
		Serosal invasion		Breast cancer
		Lymph node metastasis		Prostate cancer
		Stage		Gastric cancer
				Melanoma cancer
Cell growth	Ki-67	Lymph node metastasis	Prognosis	Prostate cancer	[104]
				Brain cancer
				Breast cancer
				Nephroblastoma
	TGF-β	Stage	Prognosis	Breast cancer	[105,106]
				Gastric cancer
Signal	PKP3	Stage	Prognosis	Gastric cancer	[109,110]
Protein/cell adhesion				Lung cancer
	E-cadherin	Invasion	Prognosis	Esophagus cancer	[111]
		Grading		Ovary cancer
		Lauren histotype		Stomach cancer
	TSP-1	Stage	Prognosis	Breast cancer	[112-115]
				Melanoma cancer
				Lung cancer
				Gastric cancer
	SPARC	Depth of invasion	Prognosis	Breast cancer	[118-120]
		Serosal invasion		Melanoma cancer
		Lymph node metastasis		Gastric cancer
		Distant metastasis		Oesophageal cancer
				Glioma cancer
Regulate cell death (apoptosis)	Bcl-2	Lymph node metastasis	Prognosis	Melanoma cancer	[121]
				Breast cancer
				Prostate cancer
				Chronic lymphocytic Leukemia lung cancer
	Bcl-6	Lymph node metastasis	Prognosis	Gastric cancer	[122]
				Pancreas cancer
				Colon cancer
				Bladder cancer
	Bax	No data	Prognosis	Gastric cancer	[123-125]
				Colon cancer
Cytokine	CXCL1/CXCR2	Lymph node metastasis	Prognosis	Gastric cancer	[128]
				Colon cancer
Vesicular transport	ARF1	Depth of invasion	Prognosis	Gastric cancer	[129]
		Serosal invasion
		Lymph node metastasis
		Distant metastasis
		Stage

HER2/neu: Human epidermal growth factor receptor 2/neu; VEGF: Vascular endothelial growth factor; hERG1: Human Ether-à-go-go-Related Gene; KLF5/IKLF/BTEB2: Krueppel-like factor 5; SATB1: Special AT-rich sequence-binding protein-1; CA IX: Carbonic anhydrase 9; MMP-2: Matrix metalloproteinase-2; HDAC: Histone deacetylases; SLP1: Antiprotease secretory leukocyte protease inhibitor; GLO1: Glyoxalase I; TGF-β: Transforming growth factor beta; PKP3: Plakophilin-3; TSP-1: Thrombospondin 1; SPARC: Secreted protein and rich in cysteine; Bcl-2: B-cell lymphoma 2; Bcl-6: B-cell lymphoma 6; Bax: Bcl-2-associated X protein; CXCL1: Chemokine (C-X-C motif) ligand 1; CXCR2: CXC chemokine receptors 2; ARF1: ADP-ribosylation factor 1.

HUMAN EPIDERMAL GROWTH FACTOR RECEPTOR 2/NEU

Human epidermal growth factor receptor 2/neu (HER2/neu) is a well-known oncogene. HER2/neu belongs to a member of the epidermal growth factor receptor family and plasma membrane-bound receptor tyrosine kinases. Overexpression of this oncogene has been shown to play an important role in the tumorigenesis of aggressive types of breast cancer. Thus, the protein has become an important tumor marker and target of therapy for over 30% of patients with breast cancer. It is positively associated with increased breast cancer recurrence and poor prognosis. Overexpression of HER2/neu has also been reported in ovarian, stomach, uterine and endometrial cancers. Dimerization results in the autophosphorylation of tyrosine residues within the cytoplasmic domain of the receptors and initiates a variety of signaling pathways.

IHC is used to measure levels of HER2/neu protein in samples. Fluorescence in situ hybridization (FISH) assays may also be used to measure whether HER2/neu gene amplification has occurred. The extracellular domain of HER2 may be shed from the surface of tumor cells and enter the circulation. Testing serum for HER2 using enzyme-linked immunosorbent assay (ELISA) offers a method for determining HER2 levels. Changes in serum HER2 concentrations may be useful in predicting response to trastazumab treatment[58-63].

VASCULAR ENDOTHELIAL GROWTH FACTOR

Vascular endothelial growth factor (VEGF) is an angiogenic factor. It is a signaling protein secreted by many solid tumors. VEGF overexpression has been reported in many human cancer cells including breast, colon, gastric, liver and lung cancers. It is also expressed in stromal cells, particularly at sites of relative hypoxia. VEGF enhances endothelial cell proliferation and migration and induces endothelial cell angiogenesis. VEGF is also called vascular permeability factor. Thus, VEGF increases vascular permeability to plasma and its proteins, a typical requirement of tumor microvasculature. IHC and ELISA are popular methods used to examine expression patterns of VEGF in serum and tissues from cancer patients[64-68].

HUMAN ETHER-À-GO-GO-RELATED GENE

Human ether-à-go-go-related gene (hERG1) belongs to the K⁺ channel family, which may regulate cell apoptosis and proliferation. Overexpression of this protein has been observed in several tumors, such as primary human endometrial, gastric and colorectal cancers. hERG1 gene expression has been used as a specific tumor marker in colon cancer, and it regulates the invasion of tumor cells[69].

KRUEPPEL-LIKE FACTOR 5 (KLF5/IKLF/BTEB2)

KLF5/IKLF/BTEB2 is a member of the Kruppel-like transcription factor family. Several members of the KLF family have been implicated in the development of human cancers[70]. KLF5 repeatedly shows genetic deletions and decreased expression in human prostate and breast cancers[71,72]. KLF5 also plays a suppressive role in tumor growth in colon cancer. However, overexpression of KLF5 activated fibroblast cell growth and tumorigenesis. KLF5 has also been reported to affect H-Ras, Wnt-1, platelet-derived growth factor A chain (PDGF), cyclin D1, PPARg and ERBB2 oncogenes[73]. Moreover, several growth factors, such as PMA, sphingosine-1-phosphate, and β-FGF, induce the expression of KLF5. Furthermore, KLF5 has been reported to promote angiogenesis in a knock-out mouse model. In a bladder cancer cell line, KLF5 was reported to inhibit cell growth. These findings suggest that the actual role of KLF5 in tumorigenesis remains unknown[74].

SPECIAL AT-RICH SEQUENCE-BINDING PROTEIN-1

Special at-rich sequence-binding protein-1 (SATB1) is an AT-rich sequence-binding protein 1 and gene regulator. It functions as a genome organizer. SATB1 is highly expressed in many types of human cancers including breast, gastric, colon and bladder carcinomas. It enables tumor growth and metastasis by altering the expression of a large number of genes[75-78].

C-MYC2

c-Myc protein is a transcription factor that activates and represses the expression of many genes. In addition, c-myc has three isoforms and is extremely influential in controlling DNA replication[79].

The MAPK/ERK pathway may activate c-myc, and c-myc activation results in many cellular effects including cell proliferation, apoptosis, differentiation and stem cell self-renewal. c-Myc is a very robust proto-oncogene, and it is very often upregulated in many cancers. c-Myc is correlated with recurrence and may be a potential prognostic factor. High levels of c-myc2 were correlated with lymph node metastasis and poor prognosis in cancers of the cervix, colon, breast, lung and stomach[80].

CARBONIC ANHYDRASE 9

Carbonic anhydrases (CAs) are a large family of zinc metalloenzymes that catalyze the reversible hydration of carbon dioxide. They are involved in some biological reactions, such as respiration, calcification, acid-base balance, bone resorption, and the formation of aqueous humor, cerebrospinal fluid (CSF), saliva, and gastric acid. CA9 is a transmembrane protein and a tumor-associated carbonic anhydrase isoenzyme. It is increased in renal cancers and hypoxic solid tumors. It is also involved in cell proliferation and tumorigenesis.

Drug compounds targeted to CA9 are in preclinical development for the treatment of kidney, colon, and cervical cancers. CA9 is one of the markers of tumor hypoxia. Moreover, recent studies assaying CA9 levels and various clinicopathological outcomes indicated that CA9 expression may also be a valuable prognostic factor of overall survival. Antibodies targeted against CA9 may be used to show hypoxic regions in many solid tumors[81-83].

MATRIX METALLOPROTEINASE-2

Matrix metalloproteinase-2 (MMP-2) is a collagenase that assists tumor growth and invasion by digesting the extracellular matrix surrounding the tissue. MMP-2 is considered a possible tumor marker for ovarian, colon, breast, bladder and gastric cancers[79,80,84,85].

HISTONE DEACETYLASES

Histone deacetylases (HDACs) promote the removal of acetyl groups from acetylated residues to release an acetate molecule. Lysine acetylation typically occurs in a number of non-histone proteins.

Mutation and/or aberrant expression of various HDACs have often been observed in cancer, making them important therapeutic targets for many human cancers. Several studies have indicated that overexpression of HDACs is present in renal, colorectal, gastric, breast, pancreatic, liver, lung and prostate cancers and in classical Hodgkin’s lymphoma and that they are associated with poor prognoses[86-92].

CYCLOOXYGENASE-2

Cyclooxygenase (COX)-2 is an enzyme responsible for inflammation and pain. COX-2 appears to be associated with cancers and irregular growths in the intestinal tract[93]. COX inhibitors have been shown to reduce the occurrence of cancers. COX-2 inhibitors are currently being studied for the treatment of breast cancer[94].

ANTIPROTEASE SECRETORY LEUKOCYTE PROTEASE INHIBITOR

Secretory leukocyte protease inhibitor (SLPI) neutralizes elastase, tryptase, and cathepsin G[95]. Several studies have indicated that SLPI may be involved in tumorigenesis[96]. Abundant expression of SLPI was noted in several cancers including ovarian, cervical and gastric cancers[97].

GLYOXALASE I

Glyoxalase I (GLO1) is an essential component in pathways leading to the detoxification of methylglyoxal (MG). Over-expression of GLO1 has been reported in several cancers, such as colon, breast, prostate, gastric cancers and melanoma[98-103].

KI-67

Ki-67 is a nuclear protein that is a cellular marker of cell proliferation. It always correlates with cell proliferation. During interphase, Ki-67 may be exclusively detected within the cell nucleus, and when the cell enters mitosis, the majority of Ki-67 relocates to the surface of the chromosomes. Ki-67 is present during all active phases of the cell cycle (G₁, S, G₂, and mitosis), but is absent in resting cells (G₀). Thus, Ki-67 is a good marker for determining cell growth rate. Ki-67-positive tumor cells are often correlated with advanced tumors including carcinomas of the prostate, brain, breast and nephroblastomas. With respect to these tumors, survival and tumor recurrence have been positively correlated with Ki-67 IHC staining[104].

TRANSFORMING GROWTH FACTOR-β

Transforming growth factor-β (TGF-β) regulates cellular proliferation, differentiation and apoptosis. TGF-β is a multifunctional cytokine, and increased or decreased production has been linked to many cancers including breast cancer and GC[105].

TGF-β is a pluripotent cytokine with diverse effects on normal physiology and a role in both normal mammary gland development and the progression of breast cancer. During early stages of cancer, TGF-β acts as a tumor suppressor, whereas in later stages, when tumor cells become resistant to growth inhibition by TGF-β, it acts as a tumor promoter. Thus, TGF-β as a tumor marker is useful primarily during the follow-up of cancer patients and particularly in monitoring advanced disease[106].

PLAKOPHILIN-3

Plakophilin-3 (PKP3) is a member of the armadillo (ARM)/plakophilin families. ARM-related proteins physically localize to cell desmosomes and nuclei and are involved in linking cadherins to intermediate filaments in the cytoskeleton. PKP3 may play a role as in cellular desmosome-dependent adhesion and signaling pathways[107,108]. Several publications have reported that genetic aberrations in members of the ARM-protein family including plakoglobin (PKGB), h-catenin (CTNNB1), and adenomatous polyposis coli (APC) promote tumor progression. However, with regard to the role(s) of the subfamily member PKP3 during carcinogenesis, evidence shows that PKP3 is a potential prognostic marker and that it shows clinicopathological correlation with lung cancer and GC[109,110].

E-CADHERIN

E-cadherin is a calcium-regulated transmembrane cell-cell adhesion protein that is expressed in most normal epithelial tissues. Selective loss of E-cadherin can cause dedifferentiation and invasiveness in human carcinomas, suggesting that E-cadherin may play a role as a tumor suppressor. Reduced expression of E-cadherin has been observed in aggressive tumors of the esophagus, ovary, and stomach[111].

THROMBOSPONDIN 1

Thrombospondin 1 (TSP-1) is an extracellular matrix glycoprotein that regulates cell adhesion, cell motility, angiogenesis, and cell growth. TSP1 is down-regulated in many tumors including breast, melanoma, lung and gastric cancers[112-115].

SECRETED PROTEIN AND RICH IN CYSTEINE

Secreted protein and rich in cysteine (SPARC) is a member of the matricellular protein family. SPARC may regulate cell-matrix interactions and cell function without participating in the structural scaffold of the extracellular matrix[116,117]. Recently, SPARC over-expression has been reported in breast, melanoma, gastric, esophageal cancers and gliomas[118-120].

B-CELL LYMPHOMA 2/B-CELL LYMPHOMA 6/BCL-2-ASSOCIATED X PROTEIN

B-cell lymphoma 2 (Bcl-2) is a member of the Bcl-2 family of regulator proteins that either induce or inhibit cell apoptosis. Bcl-2 is an important anti-apoptotic oncoprotein. Mutation of the Bcl-2 gene causes cancers including melanoma, chronic lymphocytic leukemia, breast, gastric, prostate, and lung cancers[121]. Bcl-2 is a commonly used IHC marker, and it has been used in the classification of lymphoid processes and the identification of specific epithelial tumors.

Bcl-6 is also a biomarker of gastric, pancreas, colon and bladder cancers[122].

Bcl-2-associated X protein (BAX) is a member of the Bcl-2 gene family. BAX is a pro-apoptotic factor based on its binding to Bcl-2. It is also a prognostic marker of colon and bladder cancers[123-125].

CXCL1/CXCR2

CXCL1 is thought to be a chemo-attractant for neutrophils and lymphocytes[126,127]. However, it was observed that neutrophil infiltration into tumors was dependent on CXCR2 signaling. Moreover, the expression of CXCL1/CXCR2 was significantly associated with GC progression. This evidence is consistent with the previous observation the CXCL1/CXCR2 signaling pathway was important in the metastasis of several cancers including breast, gastric and colon cancers[128].

ADP-RIBOSYLATION FACTOR 1

ADP-ribosylation factor 1 (ARF1) is a member of the small GTPase family. ARF1 is involved in many cellular processes including cell differentiation, proliferation, necrosis, apoptosis, and inter- and intracellular signaling[129].

CONCLUSION

There are no current excellent tumor indices or tumor markers for GC. However, it is very important to use many tumor markers in different types of cancer to screen, diagnose and stage the cancer, for evaluation of prognosis and for monitoring of recurrence following treatment. However, determining the significance of the markers and assessing and interpreting them requires an accumulation of expertise and experience. These processes cannot work when based on a single tumor marker. From the viewpoint of the clinic, it is critical to examine the performance of the marker using imaging, to select a tumor marker assay and to consider the variety of non-specific factors that may have caused the increase in the biomarker. Finally, this knowledge may lead to more accurate diagnoses and prognoses, may enable the selection of optimal and appropriate therapies and also may enable assessments of the effectiveness of treatments, with or without postoperative tracking of recurrence. Thus, tumor markers are only references; they are used by the physician for examining the cancer and cannot diagnose the disease themselves.