Low expression of hsa_circ_0006633 in human gastric cancer and its clinical significances

Abstract

Circular RNAs are new type of endogenous RNAs, which play an important role in the regulation of gene expression and indicate a great capacity in clinical diagnosis and treatments of diseases. However, the role of circular RNAs in gastric cancer remains unknown. In this study, we chose hsa_circ_0006633 as the target circular RNA and measured its levels in human gastric cancer tissues, plasma, and gastric cell lines by real-time quantitative reverse transcription polymerase chain reaction. Hsa_circ_0006633 levels at multiple stages of gastric tumorigenesis were then explored, and its relationships with clinicopathological features were analyzed as well. We found that the expression levels of hsa_circ_0006633 in four gastric cancer cell lines, HGC-27, SGC-7901, MGC-803, and AGS, were downregulated than those in normal gastric mucosal epithelial cell line GES-1. Then, we further detected that it was downregulated in 79.2% (76/96) gastric cancer tissues compared with the adjacent non-tumorous tissues. The lower expression of hsa_circ_0006633 was associated with cancer distal metastasis (p = 0.037) and tissue carcinoembryonic antigen level (p = 0.041). In addition, hsa_circ_0006633 expression was significantly decreased in gastritis and dysplasia tissues comparing with the healthy control. Moreover, plasma hsa_circ_0006633 levels were significantly increased in gastric cancer compared with healthy control. Our data imply that hsa_circ_0006633 may play an important role in gastric carcinogenesis and is also a potential biomarker for screening gastric cancer.

Keywords

Gastric cancer circular RNA Hsa_circ_0006633 clinical significance

Introduction

Gastric cancer is one of the most common malignant tumors worldwide and the second leading cause of global cancer deaths in the world, especially in Eastern Asian countries.^1,2 As gastric cancer often remains symptomless, it may be detected in late stage, and the 5-year survival rate remains 4%–27% from developing to developed countries, respectively.³ While the 5-year survival rate can be more than 90% after the medical measures if detected in the early stage of gastric cancer.⁴ Improved outcomes can be achieved with the early recognition and aggressive managements if detailed molecular characterization of tumors can be discovered.⁵ Therefore, it is necessary to improve the diagnosis, preferably non-invasive and cost-effective methods.

Nowadays, with the development of RNA sequencing (RNA-seq) and bioinformatic analysis, circular RNA (circRNA) has become a hot spot in the RNA field as a novel kind of RNA.⁶ Some characteristics such as diversity, abundance, and conservation indicate that circRNAs have biological functions.⁷ With their dynamic and cell-type-specific expression patterns during development, a variety of functions have been proposed for circRNAs.^6,7 CircRNAs also play specific roles in a wide variety of cancers and in regulating the biological behavior of cancers, providing a novel potential perspective in cancer treatment.⁸

Based on our previous microarray screening results of the differential expression of circRNAs in gastric cancer tissues comparing with adjacent non-tumorous tissues (GEO No. GSE89143; https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE89143), here we choose hsa_circ_0006633 as the target in the study. Among the dysregulated circRNAs, hsa_circ_0006633 is one of the most downregulated circRNAs in gastric cancer tissues. Its gene is located at chromosome 1: 59805629-59844509, and the spliced sequence length is 353 nt. Our data here showed that hsa_circ_0006633 may play an important role in the tumorigenesis of gastric cancer.

Materials and methods

Specimens and clinical data collection

All 338 samples were collected from the Affiliated Hospital of Ningbo University School of Medicine, China, from November 2012 to June 2016. The 96 gastric cancer tissues and their adjacent non-tumorous tissues 5 cm from the edge of the tumor were obtained from surgical operations. Another 35 human healthy gastric mucosa, 51 gastritis mucosa, and 20 gastric dysplasia tissues were gained from biopsy specimens. All tissues were preserved in RNA-fixer Reagent (Bioteke, Beijing, China) at −80°C until use. Plasma samples were obtained from 20 healthy volunteers and 20 gastric cancer patients and stored at −80°C until RNA extraction. No patient or healthy volunteer received local or systemic treatment before the upper gastrointestinal endoscopy examination or surgical excision. The final diagnosis was histopathologically confirmed. Tumors were staged according to the tumor–node–metastasis (TNM) stage system of the international Union Against Cancer (7th edition). Histological grade was assessed following the National Comprehensive Cancer Network (NCCN) clinical practice guideline of oncology (V.1, 2011). The experimental operators were blind to the clinical data, which were collected by doctors. The study was approved by the Human Research Ethics Committee of Ningbo University, and all patients agreed to participate in the study and written informed consent was obtained from all subjects.

Cell cultures

Human gastric epithelial cell line GES-1 and four gastric cancer cell lines, HGC-27, SGC-7901, MGC-803, and AGS, were obtained from Shanghai Institute of Biochemistry and Cell Biology, Chinese Academy of Sciences (Shanghai, China). Cells were cultured with RPMI 1640 Medium (Invitrogen, Grand Island, NY, USA) consists of 50 U/mL penicillin, 10% fetal bovine serum, and 50 U/mL streptomycin in culture flasks at 37°C in a humidified atmosphere of 5% CO₂ as previous report.⁹

Total RNA preparation and qRT-PCR detection

We first extracted total RNA from specimens by TRIzol reagent (Ambion, Carlsbad, CA, USA) and then reverse transcribed it into cDNA by GoScript Reverse Transcription (RT) System (Promega, Madison, WI, USA) following the manufacturer’s instructions. Real-time quantitative reverse transcription polymerase chain reaction (qRT-PCR) was achieved using GoTaq qPCR Master Mix (Promega) on an Mx3005P real-time PCR System (Stratagene, La Jolla, CA, USA). Divergent primers rather than the more commonly used convergent primers of hsa_circ_0006633 were designed.¹⁰ Primers for hsa_circ_0006633 and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) were synthesized by Sangon Biotech (Shanghai, China). The sequences of the PCR primers for GAPDH and hsa_circ_0006633 were as follows: 5′-TCGACAGTCAGCCGCATCTTCTTT-3′ and 5′-ACCAAATCCGTTGACTCCGACCTT-3′ for GAPDH; 5′-CTCCCGGACTTCTTATCGTGG-3′ and 5′-CGATGGTCCAGCCACATGAT-3′ for hsa_circ_0006633. The data were analyzed through the ΔC_t method,¹¹ and all the results were expressed as mean ± standard deviation (SD) of three independent experiments. Higher ΔCt means lower expression of hsa_circ_0006633. The conditions of thermal cycling were as follows: 95°C at 5 min for a hot-start, then 40 cycles at 94°C for 15 s, 56°C for 30 s, and 72°C for 30 s. The cycle threshold (C_t) values were recorded for both hsa_circ_0006633 and GAPDH, which is used as a control.

Immunohistochemical analysis of tissue carcinoembryonic antigen and CA19-9

We incubated the paraffin tissue sections in primary anti-carcinoembryonic antigen (CEA) or anti-carbohydrate antigen 19-9 (CA19-9) (Dako, Glostrup, Denmark) for 1 h at room temperature, and then, the tissues were incubated in diaminobenzidine (DAB; Dako) for color development after incubation with broad-spectrum second antibody K5007 (Dako).¹² The standard for determination of the results was according to the 2010 American Society of Clinical Oncology (ASCO)/the College of American Pathologists (CAP) guideline.¹²

Statistical analysis

All statistical analyses were performed by Statistical Program for Social Sciences (SPSS) 20.0 software (SPSS, Chicago, IL, USA), GraphPad Prism 5.0 (GraphPad Software, La Jolla, CA, USA), and SigmaPlot 10.0 (SigmaPlot Software, La Jolla, CA, USA). We used Student’s t test, one-way analysis of variance (ANOVA) test, and rank-sum test as appropriate. Statistical significance was accepted at p < 0.05.

Results

Expression of hsa_circ_0006633 in gastric cancer cell lines and tissues

We first sequenced the qRT-PCR product, which was coincident with that in circBase (http://circbase.org/; Figure 1). Then, we found that the expression levels of hsa_circ_0006633 in four gastric cancer cell lines, HGC-27, SGC-7901, MGC-803, and AGS, were significantly downregulated than those in normal gastric epithelial cell line GES-1 (Figure 2).

Figure 1.

DNA sequencing results of hsa_circ_0006633 in gastric tissues. The qRT-PCR products of tissue hsa_circ_0006633 were sequenced by Qingkezixi Biotech (Hangzhou) Co., Ltd.

Figure 2.

Relative expression of hsa_circ_0006633 in gastric cancer cell lines and normal gastric epithelial cell line. Hsa_circ_0006633 expression levels in four gastric cancer cell lines, HGC-27, SGC-7901, MGC-803, and AGS, and normal gastric epithelial cell line GES-1 were determined by qRT-PCR. Data are expressed as means ± SD of three independent experiments. Asterisks indicate p values that are significantly different comparing with those in GES-1 (*p < 0.05, **p < 0.01).

We then explored hsa_circ_0006633 expression levels in gastric cancer tissues. Its level in gastric cancer tissues was significantly lower than that in corresponding non-tumorous tissues (p < 0.001, Figure 3). Its expression levels were significantly downregulated in 79.2% (76/96) gastric cancer tissues compared with the adjacent normal tissues (Figure 3(c)). To explore its diagnostic values, a receiver operating characteristic (ROC) curve was established. The area under the ROC curve was 0.741 (Figure 3(d)). When the cutoff value was 8.17, its sensitivity and specificity were 0.60 and 0.81, respectively.

Figure 3.

Hsa_circ_0006633 expression levels in gastric cancer tissues. (a) The expression levels of hsa_circ_0006633 in cancer tissues (n = 96) and adjacent normal tissues (n = 96). Higher ΔC_t value indicates lower expression (***p < 0.001). (b) The expression levels of hsa_circ_0006633 are significantly lower than those in corresponding non-tumorous tissues. (c) The expression level of hsa_circ_0006633 was significantly downregulated in 79.2% (76/96) gastric cancer tissues compared with the adjacent normal tissues. (d) ROC curve of hsa_circ_0006633 in differentiating gastric cancer tissues from controls. The area under the curve was up to 0.741.

We further detected the hsa_circ_0006633 expression levels in different stages of gastric cancer development. As shown in the Figure 4, hsa_circ_0006633 expression levels were significantly decreased in gastritis, dysplasia, and cancer tissues comparing with healthy control; meanwhile, it was significantly downregulated in gastritis tissues comparing the cancer tissues (Figure 4). There were no significant differences of hsa_circ_0006633 expression levels between dysplasia and cancer tissues.

Figure 4.

Differences of hsa_circ_0006633 expression levels among different stages of gastric mucosa. The hsa_circ_0006633 expression levels in healthy gastric mucosa (n = 35), gastritis (n = 51), dysplasia (n = 20), and gastric cancer (n = 96) tissues were determined by qRT-PCR. Higher ΔC_t value indicates lower expression (**p < 0.01, ***p < 0.001).

Relationships between hsa_circ_0006633 expression levels and clinicopathological factors

Based on the above findings, we analyzed the association between hsa_circ_0006633 expression levels and clinicopathological characteristics of patients with gastric cancer. As shown in Table 1, low expression of hsa_circ_0006633 in human gastric cancer tissues was associated with distal metastasis (p = 0.037) and tissue CEA level (p = 0.041). However, its levels were not significantly associated with other clinicopathological characteristics such as age, gender, diameter, stage, pathological diagnosis, invasion, and CA19-9 levels.

Table 1.

Relationship of Hsa_circ_0006633 expression levels (ΔC_t) in cancer tissues with clinicopathological factors of patients with gastric cancer..

Characteristics	No. of case (%)	Mean ± SD	p value
Age (years)
≥60	61 (63.5)	9.194 ± 1.518	0.327
<60	35 (36.5)	8.869 ± 1.613
Gender
Male	65 (67.7)	9.050 ± 1.450	0.815
Female	31 (32.3)	9.130 ± 1.775
Tumor location
Sinuses ventriculi	49 (51.1)	9.279 ± 1.547	0.116
Cardia	10 (10.4)	9.108 ± 1.156
Corpora ventriculi	25 (26.0)	8.450 ± 1.332
Others	12 (12.5)	9.523 ± 2.033
Diameter (cm)
≥5	47 (49.0)	9.126 ± 1.772	0.761
<5	49 (51.0)	9.028 ± 1.327
Differentiation
Well	12 (12.5)	9.103 ± 1.550	0.235
Moderate	47 (49.0)	9.328 ± 1.455
Poor	37 (38.5)	8.746 ± 1.652
Stage
Early	24 (25.0)	9.186 ± 1.700	0.691
Advanced	72 (75.0)	9.039 ± 1.512
Borrmann type
I & II	19 (26.4)	8.956 ± 1.692	0.783
III & IV	53 (73.6)	9.069 ± 1.458
Pathological diagnosis
Signet ring cell cancer	15 (15.6)	8.760 ± 1.576	0.394
Adenocarcinoma	81 (84.4)	9.134 ± 1.552
Invasion
T₁ & T₂	36 (37.5)	9.217 ± 1.633	0.492
T₃ & T₄	60 (62.5)	8.991 ± 1.511
Lymphatic metastasis
N₀	38 (39.6)	8.837 ± 1.565	0.224
N_1-3	58 (60.4)	9.232 ± 1.538
Distal metastasis
M₀	82 (85.4)	9.212 ± 1.595	0.037
M₁	14 (14.6)	8.279 ± 0.982
Venous invasion
Absent	53 (55.2)	8.983 ± 1.674	0.519
Present	43 (44.8)	9.190 ± 1.401
Perineural invasion
Absent	47 (49.0)	9.250 ± 1.613	0.284
Present	49 (51.0)	8.909 ± 1.490
CEA (tissue)
Positive	74 (77.1)	9.222 ± 1.563	0.041
Negative	22 (22.9)	8.384 ± 1.446
CA19-9 (tissue)
Positive	54 (56.3)	9.279 ± 1.465	0.147
Negative	42 (43.7)	8.815 ± 1.640

CEA: carcinoembryonic antigen; SD: standard deviation.

Detection of hsa_circ_0006633 level in human plasma

As human plasma is the main material used in the screening of cancers, we wondered whether plasma hsa_circ_0006633 levels were also altered in gastric cancer patients. We detected plasma hsa_circ_0006633 levels in 20 healthy volunteers and 20 gastric cancer patients by qRT-PCR method. Our data showed that hsa_circ_0006633 levels were significantly increased in gastric cancer patients’ plasma compared with healthy control (Figure 5).

Figure 5.

Detection of hsa_circ_0006633 level in human plasma. Plasma hsa_circ_0006633 levels in healthy volunteers (n = 20) and gastric cancer patients (n = 20) were detected by qRT-PCR method. Lower ΔC_t value indicates higher level (*p < 0.05).

Discussion

CircRNAs were once thought to be the byproducts of splicing errors caused by low abundance and the technological limitations, while they have been received great attention recent years with the development of RNA-seq and bioinformatic analysis.¹³ Unlike linear RNA, circRNA forms covalently closed-loop structure by jointing 3′ and 5′ ends together via exon circularization or intron circularization.¹³ CircRNAs are abundant, multiple, and stable in creatures.^7,14 Many researchers have found that circRNAs play an important part in the regulation of gene expression.⁶ It is widely accepted that some RNAs regulate with microRNA (miRNA) response elements (MREs) and works as “competing endogenous RNA (ceRNA)” hypothesis.^15,16 Recent studies have showed that circRNAs can work as a miRNA sponge. For example, ciRS-7 (circRNA sponge for miR-7) containing approximately 70 binding sites acts as a miR-7 sponge, which can suppress the miR-7 activity and thus increase levels of miR-7 targets; sex-determining region Y (Sry) works as a miR-138 sponge.¹⁷

In addition, circRNAs have been implicated in many diseases such as Parkinson disease, Alzheimer’s disease, atherosclerosis risk, and prion diseases.^18–21 Increasing studies have showed that circRNAs play an important role in tumorigenesis and tumor progression.²² In colorectal cancer, there exists a negative correlation of global circRNA abundance and proliferation.²³ Hsa_circ_0001649 was downregulated in hepatocellular carcinoma.²⁴ Hsa_circ_100855 and hsa_circ_104912 were dysregulated in laryngeal squamous cell cancer (LSCC) tissues and associated with some clinicopathological factors.²⁵ In gastric cancer, lower expression of hsa_circ_002059 expression was first found to be significantly correlated with distal metastasis, TNM stage, gender, and age.²⁶ In this study, we found that expression levels of hsa_circ_0006633 in four gastric cancer cell lines, AGS, HGC-27, MGC-803, and SGC-7901, were significantly downregulated than those in normal gastric epithelial GES-1 cells (Figure 2). Moreover, it was also significantly downregulated in gastric cancer tissues than those in paired adjacent non-tumorous tissues (Figure 3). The data realized that hsa_circ_0006633 may play an important role in the carcinogenesis of gastric cancer.

The occurrence of gastric cancer is the result of multi-stage and multi-factors.^2,27 It is a progressive process of the gradual emergency of gastritis, gastric atrophy, intestinal metaplasia, and dysplasia, eventually leading to the occurrence of gastric cancer.²⁸ The progression of gastric cancer is mainly caused by microbial pathogens and is closely related to host inflammatory factors; during this process, the altered microenvironment may cause random mutations to occur in gastric cells, which inevitably leads to carcinoma.^2,29 Increasing evidence has revealed that several proinflammatory cytokinesgenes, such as those for tumor necrosis factor (TNF) and interleukin-1 (IL-1), may increase the risk of gastric cancer.³⁰ In this study, we made further explorations of hsa_circ_0006633 expression in multiple stages of tissues during the gastric cancer progression. We discovered that hsa_circ_0006633 expression was significantly decreased in gastritis, dysplasia, and gastric cancer tissues (Figure 4). Since mucosal inflammation is an initiation factor eventually resulting in gastric cancer,^2,28 we found that the expression of hsa_circ_0006633 in gastritis was significantly lower than that in gastric cancer tissues (Figure 4), indicating that hsa_circ_0006633 may be a dysregulative molecule underlying gastric benign lesions.

Distal metastasis is a crucial factor in evaluating the prognosis of gastric cancer. A study indicated that patients with both serum CEA and CA19-9 positive significantly correlated higher frequencies of lymph node metastasis, deeper invasion by the tumor, and higher rates of hepatic metastasis.³¹ Our research found that the lower expression of hsa_circ_0006633 was significantly correlated with distal metastasis and tissue CEA level (Table 1). Moreover, hsa_circ_0006633 can exist in human plasma. Plasma hsa_circ_0006633 levels were significantly increased in gastric cancer patients compared with healthy control. This means that hsa_circ_0006633 is also a potential biomarker for screening gastric cancer.

In conclusion, our data implied that hsa_circ_0006633 expression was not only significantly decreased in gastric cancer tissues and its cell lines but also decreased in dysplasia. The downregulation expression of hsa_circ_0006633 was associated with distal metastasis and tissue CEA level. Moreover, compared with healthy control, plasma hsa_circ_0006633 levels were significantly increased in gastric cancer patients. Our findings suggest that hsa_circ_0006633 may play a crucial role during gastric cancer occurrence and is also a potential biomarker for screening gastric cancer.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethical approval

This study was approved by the Human Research Ethics Committee of Ningbo University School of Medicine.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by grants from the Social Development Research Project of Ningbo (No. 2016C51015), the Medical Research Project of the Affiliated Hospital of Ningbo University School of Medicine (No. XYY16022), the Foundation of Zhejiang Key Laboratory of Pathophysiology (No. 201601), the Medical and Health Research Project of Zhejiang Province (No. 2017KY598), Applied Research Project on Nonprofit Technology of Zhejiang Province (Nos 2016C33177 and 2017C35004), and the K. C. Wong Magna Fund in Ningbo University.

References

Khazaei

Rezaeian

Soheylizad

. Global incidence and mortality rates of stomach cancer and the human development index: an ecological study. Asian Pac J Cancer Prev 2016; 17(4): 1701–1704.

Yari

Abiri

Aryan

. Loop-mediated isothermal amplification as a fast noninvasive method of helicobacter pylori diagnosis. J Clin Lab Anal 2016; 30(5): 464–470.

Eybpoosh

Talebkhan

Saberi

. Age-specific gastric cancer risk indicated by the combination of helicobacter pylori sero-status and serum pepsinogen levels. Iran Biomed J 2015; 19(3): 133–142.

Song

Qiao

Gao

. A comparison of endoscopic submucosal dissection (ESD) and radical surgery for early gastric cancer: a retrospective study. World J Surg Oncol 2015; 13(1): 309.

Shao

Chen

Jiang

. Low expression of lncRNA-HMlincRNA717 in human gastric cancer and its clinical significances. Tumour Biol 2014; 35(10): 9591–9595.

Barrett

Salzman

. Circular RNAs: analysis, expression and potential functions. Development 2016; 143(11): 1838–1847.

Lasda

Parker

. Circular RNAs: diversity of form and function. RNA 2014; 20(12): 1829–1842.

Zhao

Shen

. Circular RNA participates in the carcinogenesis and the malignant behavior of cancer. RNA Biol. 2015; 14(3): 1–8.

Shao

. LncRNA-RMRP promotes carcinogenesis by acting as a miR-206 sponge and is used as a novel biomarker for gastric cancer. Oncotarget 2016; 7(25): 37812–37824.

10.

Jeck

Sharpless

. Detecting and characterizing circular RNAs. Nat Biotechnol 2014; 32(5): 453–461.

11.

Yang

Shao

Zhu

. Using gastric juice lncRNA-ABHD11-AS1 as a novel type of biomarker in the screening of gastric cancer. Tumour Biol 2016; 37(1): 1183–1188.

12.

Terada

. An immunohistochemical study of primary signet-ring cell carcinoma of the stomach and colorectum: III. Expressions of EMA, CEA, CA19-9, CDX-2, p53, Ki-67 antigen, TTF-1, vimentin, and p63 in normal mucosa and in 42 cases. Int J Clin Exp Pathol 2013; 6(4): 630–638.

13.

Salzman

. Circular RNA expression: its potential regulation and function. Trends Genet 2016; 32(5): 309–316.

14.

Suzuki

Tsukahara

. A view of pre-mRNA splicing from RNase R resistant RNAs. Int J Mol Sci 2014; 15(6): 9331–9342.

15.

. Molecular mechanisms of long noncoding RNAs on gastric cancer. Oncotarget 2016; 7(8): 8601–8612.

16.

Xia

Chen

Jiang

. Long noncoding RNA FER1L4 suppresses cancer cell growth by acting as a competing endogenous RNA and regulating PTEN expression. Sci Rep 2015; 5: 13445.

17.

Hansen

Jensen

Clausen

. Natural RNA circles function as efficient microRNA sponges. Nature 2013; 495(7441): 384–388.

18.

Hansen

Kjems

Damgaard

. Circular RNA and miR-7 in cancer. Cancer Res 2013; 73(18): 5609–5612.

19.

Lukiw

. Circular RNA (circRNA) in Alzheimer’s disease (AD). Front Genet 2013; 4(4): 307.

20.

Burd

Jeck

Liu

. Expression of linear and novel circular forms of an INK4/ARF-associated non-coding RNA correlates with atherosclerosis risk. PLoS Genet 2010; 6(12): e1001233.

21.

Satoh

Yamamura

. Gene expression profile following stable expression of the cellular prion protein. Cell Mol Neurobiol 2004; 24(6): 793–814.

22.

Zhang

. Noncoding RNAs in gastric cancer: research progress and prospects. World J Gastroenterol 2016; 22(29): 6610–6618.

23.

Bachmayr-Heyda

Reiner

Auer

. Correlation of circular RNA abundance with proliferation-exemplified with colorectal and ovarian cancer, idiopathic lung fibrosis, and normal human tissues. Sci Rep 2015; 5: 8057.

24.

Qin

Liu

Huo

. Hsa_circ_0001649: a circular RNA and potential novel biomarker for hepatocellular carcinoma. Cancer Biomark 2016; 16(1): 161–169.

25.

Xuan

Zhou

. Circular RNA: a novel biomarker for progressive laryngeal cancer. Am J Transl Res 2016; 8(2): 932–939.

26.

Chen

. Using circular RNA as a novel type of biomarker in the screening of gastric cancer. Clin Chim Acta 2015; 444: 132–136.

27.

Sun

Yang

. Roles of long noncoding RNAs in gastric cancer and their clinical applications. J Cancer Res Clin Oncol 2016; 142(11): 2231–2237.

28.

Correa

. Human gastric carcinogenesis: a multistep and multifactorial process – First American Cancer Society Award Lecture on Cancer Epidemiology and Prevention. Cancer Res 1992; 52(24): 6735–6740.

29.

Liu

Liang

. Inflammatory microenvironment contributes to epithelial-mesenchymal transition in gastric cancer. World J Gastroenterol 2016; 22(29): 6619–6628.

30.

Tsujimoto

Ono

Ichikura

. Roles of inflammatory cytokines in the progression of gastric cancer: friends or foes? Gastric Cancer 2010; 13(4): 212–221.

31.

Ishigami

Natsugoe

Hokita

. Clinical importance of preoperative carcinoembryonic antigen and carbohydrate antigen 19-9 levels in gastric cancer. J Clin Gastroenterol 2001; 32(1): 41–44.