Brief Article
Copyright ©2011 Baishideng Publishing Group Co., Limited. All rights reserved.
World J Gastroenterol. Sep 28, 2011; 17(36): 4113-4117
Published online Sep 28, 2011. doi: 10.3748/wjg.v17.i36.4113
Role of cyclooxygenase-2 gene polymorphisms in pancreatic carcinogenesis
Renata Talar-Wojnarowska, Anita Gasiorowska, Marek Olakowski, Pawel Lampe, Beata Smolarz, Hanna Romanowicz-Makowska, Ewa Malecka-Panas
Renata Talar-Wojnarowska, Anita Gasiorowska, Ewa Malecka-Panas, Department of Digestive Tract Diseases, Medical University, 22 Kopcinskiego, 90-153 Lodz, Poland
Marek Olakowski, Paweł Lampe, Department of Digestive Tract Surgery, Silesian Medical University, 14 Medykow, 40-752 Katowice, Poland
Beata Smolarz, Hanna Romanowicz-Makowska, Laboratory of Molecular Genetics, Institute of Polish Mother’s Memorial Hospital, 281/289 Rzgowska, 93-338 Lodz, Poland
Author contributions: Talar-Wojnarowska R and Gasiorowska A provided the protocols and performed the research; Olakowski M and Lampe P provided the collection of human material and analyzed the data; Smolarz B and Romanowicz-Makowska H carried out the molecular analysis; and Talar-Wojnarowska R and Malecka-Panas E designed the study and wrote the paper.
Correspondence to: Renata Talar-Wojnarowska, MD, PhD, Department of Digestive Tract Diseases, Medical University of Lodz, Kopcinskiego 22, 90-153 Lodz, Poland. renata.talar-wojnarowska@umed.lodz.pl
Telephone: +48-42-6786480  Fax: +48-42-6786480
Received: December 20, 2010
Revised: March 26, 2011
Accepted: April 2, 2011
Published online: September 28, 2011

Abstract

AIM: To evaluate the clinical significance of -765G/C and -1195G/A cyclooxygenase-2 (COX-2) gene polymorphisms in patients with pancreatic cancer (PC).

METHODS: The study included 201 patients: 85 with PC and 116 healthy controls. -765G/C and -1195G/A COX-2 gene polymorphisms were studied in DNA isolated from blood samples. The associations of the analyzed genotypes and clinical data at diagnosis were evaluated.

RESULTS: We found an increased frequency of the homozygous -1195AA COX-2 genotype in patients with PC (53.7%) compared with the control group (21%) (P < 0.01). In contrast, the distribution of genotype and allele frequencies of the -765G/C COX-2 polymorphism in the PC patients were not different from those in control groups. A correlation between presence of homozygous -1195AA COX-2 genotype and tumor size > 3 cm was observed (P < 0.05). Analyzed polymorphisms were unrelated to the patients’ sex and age, nor to the presence of regional or distant metastases.

CONCLUSION: These preliminary results indicate that the -1195G/A COX-2 polymorphism may play an important role in PC prognosis and carcinogenesis.

Key Words: Cyclooxygenase-2, Polymorphisms, Pancreatic cancer, Carcinogenesis



INTRODUCTION

Cyclooxygenase-2 (COX-2), also known as prostaglandin endoperoxide synthase, is a key enzyme in the arachidonic acid pathway, initiating the synthesis of biologically important prostanoids and eicosanoids. COX-2 is involved in many biologic processes, such as cell proliferation, invasion, angiogenesis and inhibition of apoptosis, which are all relevant to cancer development and progression[1]. COX-2 is expressed under certain extracellular or intracellular stimuli, such as mitogens, growth factors, hormones, infectious agents and proinflammatory cytokines[1,2]. Numerous studies have demonstrated increased expression of COX-2 in various cancers, including pancreatic tumors[3-6]. In several studies, the overexpression of COX-2 in pancreatic cancer (PC) cells has been shown to be an independent prognostic factor and to increase the clinical aggressiveness of this disease[3,4,7].

The COX-2 gene was demonstrated to be genetically polymorphic, which may affect the expression or activity of this enzyme and consequently contribute to variation in individual susceptibility to cancer through aberrant arachidonic acid metabolism. Genetic variants represented by single nucleotide polymorphisms (SNPs) of the COX-2 gene, among them -765G/C and -1195G/A in the promoter region, have been identified. The functional effects of these SNPs have been recently confirmed[8-10]. The -765G allele is associated with heightened COX-2 transcription by creating a transcriptional factor c-MYB-binding site and with significantly higher promoter activity compared with -765C allele[8,9]. In the study of Zhang et al[10], the -1195A-containing haplotypes also had significantly increased COX-2 messenger RNA levels in esophageal tissues compared with the -1195G-containing counterparts.

COX-2 has been the object of prevention/intervention strategies in many clinical trials, including being a potential therapeutic target for chemoprevention and therapy of PC[1,2,11]. Selective inhibition of COX-2 results in variable responses in individual patients. Information regarding the functional significance of COX-2 polymorphisms with risk-modulating ability would have significant implications, not only for risk identification, but also for pharmacological management of the disease.

The purpose of this study was to evaluate the clinical significance of -765G/C and -1195G/A COX-2 gene polymorphisms in patients with PC.

MATERIALS AND METHODS

The study included 201 Caucasian patients: 85 with PC (41 men and 44 women, aged 44-84 years) and 116 gender- and age-matched healthy volunteers. Analyzed patients were hospitalized in the Department of Digestive Tract Diseases, Medical University of Lodz Hospital or in the Department of Digestive Tract Surgery of Silesian Medical University in Katowice between 2004 and 2009. Only patients with a confirmed pathological diagnosis of ductal pancreatic adenocarcinoma were included in the study. The pathological diagnosis was confirmed after surgical treatment or after pancreatic tissue biopsy in patients who qualified for palliative chemotherapy. Twenty-nine patients (34.1%) with PC underwent Whipple resection or distal pancreatectomy, 33 patients (38.8%) underwent palliative surgery and 23 patients (27.1%) underwent palliative chemotherapy and/or endoscopic treatment. Tumor grade was classified into G1 (well differentiated), G2 (moderately differentiated) and G3 (poorly differentiated). The study protocol was approved by the ethical committee of Lodz Medical University.

Peripheral venous blood samples were obtained from all analyzed patients at the time of hospital admission. -765G/C and -1195G/A COX-2 gene polymorphisms were studied in DNA isolated from blood samples using the QIAmp DNA Mini Kit (Qiagen). Polymerase chain reaction (PCR) products for the COX-2 variants were analyzed by the restriction fragment length polymorphism method. The primers used to amplify the COX-2 promoter region were 5’-TAT TAT GAG GAG AAT TTA CCT TTC GC-3’ and 5’-GCT AAG TTG CTT CAA CAG AAG AAT-3’ for the -765G/C variant; and 5’-CCC TGA GCA CTA CCC ATG AT-3’ and 5’-GCC CTT CAT AGG AGA TAC TGG-3’ for the -1195G/A polymorphism. PCR amplification was per-formed in a final volume of 25 μL containing 30-100 ng of DNA, 10 mmol/L Tris-HCl (pH 8.3), 4 μL of 25 mmol/L MgCl2, 50 mmol/L KCl, 0.5 μL dNTP (10 mmol/L), each primer at 1.0 μmol/L and 1.0 unit of Taq polymerase (BIOKOM, Takara, Japan) in a GeneAmp PCR system 9700 Thermocycler (Applied Biosystems).

Ten microliters of the PCR product were digested with 2 units of restriction enzymes HhaI or PvuII (BioLaps, New England) using the manufacturer’s recommended protocol. PCR products were visualized on 8% polyacrylamide gels with 10% ethidium bromide. COX-2 genotypes that could be detected were respectively: -765CC (100 bp fragment), -765GC (100 and 74 and 26 bp fragments), -765GG (74 and 26 bp fragments), -1195AA (273 bp fragment), -1195GA (273 and 220 and 53 bp fragments) and -1195GG (220 and 53 bp fragments).

The serum concentrations of CA19-9 were measured by an enzyme-linked immunoassay (DRG International, United States), according to the manufacturer’s recommendations. The associations of the analyzed genotypes and patient characteristics at PC diagnosis were evaluated. The following demographic and clinical data were analyzed: age, gender, tumor size, lymph node involvement, histological grade, CA19-9 levels, weight loss and history of smoking. The cut-off point of CA19-9 was set at 37 U/mL.

Statistics analysis

The results were analyzed using StatSoft Statistica for Windows, release 6.0 (StatSoft, Inc., Tulsa, United States). To determine differences between groups, Mann-Whitney t tests were used. Clinical significance of analyzed polymorphisms was determined using logistic regression analysis and presented in tables as odds ratios (OR) with their 95% confidence intervals. The deviations from Hardy-Weinberg equilibrium were analyzed using the χ2 test. Differences with a P value less than 0.05 were considered significant.

RESULTS

All patients involved in the study were Caucasians. Mean ages were not significantly different for patients with PC (mean 66.8 ± 4.1 years) and controls (63.1 ± 4.7 years, P > 0.05). In patients with pancreatic adenocarcinoma, the tumor size ranged from 2 cm to 7 cm (mean 3.7 ± 2.3). As for histological differentiation, 19, 29 and 35 patients were classified into G1, G2 and G3 respectively, whereas 2 patients had missing data. Lymph node metastases were observed in 39 patients with PC (45.9%) and liver metastases in 16 of them (18.8%). Serum levels of CA19-9 were higher in patients with PC compared to control group (P < 0.001; respectively 101.2 ± 21.4 U/mL vs 17.6 ± 3.2 U/mL; data shown in our previously published work[22]).

The genotype distributions of analyzed -765G/C and -1195G/A COX-2 gene polymorphisms are summarized in Table 1. We found an increased frequency of the homozygous -1195AA COX-2 genotype in patients with PC compared with control group [OR 6.48 (2.93-14.31), P < 0.01]. In contrast, the distribution of genotype and allele frequencies of the -765G/C COX-2 polymorphism in the PC patients did not differ from those in control groups (Table 1). Each of the COX-2 polymorphisms in the controls was consistent with Hardy-Weinberg equilibrium.

Table 1 Distribution of -1195 G/A and -765 G/C COX-2 genotype in the analyzed group of patients n (%).
GenotypePC patients(n = 85)Control group(n = 116)OR (95% CI)
-1195 G/A
GG13 (15.7)44 (37.9)Reference
GA26 (30.6)48 (41.4)1.83 (0.84-4.00)
AA46 (53.7)24 (21.0)6.48 (2.93-14.31)
- 765 G/C
GG47 (55.4)44 (37.9)Reference
GC27 (31.7)40 (34.5)0.63 (0.33-1.19)
CC11 (12.9)32 (27.6)0.32 (0.14-1.71)

The potential relationship between COX-2 genotype distribution and clinical data of the PC patients was investigated. COX-2 -1195AA genotype showed a significant association with tumor size > 3 cm in patients with PC (P < 0.05, Table 2). This analyzed polymorphism was unrelated to the patients’ sex and age, weight loss, history of smoking, CA19-9 levels, nor with the presence of regional or distant metastases. In contrast, the -765G/C COX-2 polymorphism was not associated with any clinical data (Table 3).

Table 2 Relationship between -1195 G/C COX-2 polymorphism and clinical data of patients with pancreatic cancer n (%).
GroupG+ allele(GA and GG)(n = 39)P valueG- allele(AA)(n = 46)P value
Age< 6520 (51.3)NS19 (41.3)NS
≥ 6519 (48.7)27 (58.7)
GenderMale18 (46.2)NS23 (50.0)NS
Female21 (53.8)23 (50.0)
Tumor size ≤ 3 cm21 (53.8)NS13 (28.3)P < 0.05
> 3 cm18 (46.2)33 (71.7)
TumorG1 + G221 (53.8)NS27 (58.7)NS
DifferentiationG317 (43.8)18 (39.2)
Lymph nodeAbsent23 (58.9)NS23 (50.0)NS
MetastasesPresent16 (41.1)23 (50.5)
Weight loss< 10 %19 (48.7NS25 (54.4)NS
≥ 10%20 (51.3)21 (45.6)
SmokingYes21 (53.8)NS19 (41.3)NS
No18 (46.2)27 (58.7)
CA19-9< 37 U/mL10 (25.6)NS13 (28.3)NS
≥ 37 U/mL29 (74.4)33 (71.7)
Table 3 Relationship between -765 G/C COX-2 polymorphism and clinical data of patients with pancreatic cancer n (%).
GroupG+ allele(GC and GG)(n = 47)P valueG- allele(CC)(n = 38)P value
Age< 6523 (48.9)NS16 (42.1)NS
≥ 6524 (51.1)22 (57.9)
GenderMale22 (46.8)NS19 (50.0)NS
Female25 (53.2)19 (50.0)
Tumor size ≤ 3 cm17 (40.4)NS15 (39.5)NS
> 3 cm28 (59.6)23 (60.5)
TumorG1 + G227 (57.4)NS21 (55.3)NS
DifferentiationG320 (40.4)16 (42.1)
Lymph nodeAbsent28 (59.8)NS18 (47.4)NS
MetastasesPresent19 (40.4)20 (52.6)
Weight loss< 10 %30 (63.8)NS21 (55.3)NS
≥ 10%17 (36.2)17 (44.7)
SmokingYes23 (48.9)NS17 (44.7)NS
No24 (51.1)21 (55.3)
CA19-9< 37 U/mL12 (25.5)NS12 (31.6)NS
≥ 37 U/mL35 (74.5)26 (68.4)
DISCUSSION

The overexpression of COX-2 has been shown to induce angiogenesis by increased synthesis of vascular endothelial growth factor and to inhibit apoptosis by activation of proto-oncogene Bcl-2[2]. It is known that the expression of COX-2 is increased in the majority of PC cells and may represent a target for adjuvant therapy of PC. However, little is known about the role of COX-2 gene polymorphisms in pancreatic carcinogenesis. We investigated the clinical significance of the -765G/C and -1195G/A COX-2 gene polymorphisms, as well as their potential association with the risk of developing PC.

In our study, the presence of the -1195AA genotype was found more frequently in patients with PC compared to the control group. Similarly, Zhao et al[8] observed that subjects carrying the COX-2 -1195A allele had significantly increased risk for developing PC compared with subjects carrying the -1195G allele. In previous studies, the association of the -1195A allele with an increased risk of lung, oral and esophageal cancers was also demonstrated[12-14]. In the study of Bi et al[13], -1195AA genotype was significantly correlated with worse overall survival (15.7 mo vs 20.2 mo, P = 0.006) and with shorter progression-free survival (9.5 mo vs 11.9 mo, P = 0.0034) in patients with unresectable locally advanced non-small cell lung cancer.

However, others authors observed opposite results. In the study of Kristinsson et al[15], the -1195GG genotype resulted in a higher risk of developing esophageal adenocarcinom. On the other hand, Pereira et al[16] ob-served that men carrying the -1195G allele appeared to have a nine-fold increased risk for colorectal cancer. Racial and ethnical differences in the studies’ populations may explain these contradictory results, because the distribution of COX-2 polymorphisms may differ considerably between populations.

The published data about clinical significance of the second analyzed polymorphism of COX-2 gene, -765G/C, are also controversial. In the study of Hoff et al[17], the -765GG genotype was present more often in patients with colorectal cancer compared to control group. Similarly, Coskunpinar et al[14] observed increased risk of lung carcinoma in Turkish patients carrying the -765G allele. In contrast, the -765C allele was associated with an increased risk for developing PC and urinary bladder cancer[8,18]. This is not in line with our findings, since we could not demonstrate a significant difference in -765G/C genotype distribution in patients with PC. Similarly, Dong et al[19] in a meta-analysis of 47 case-control studies did not find a convincing association between -765G/C COX-2 gene polymorphism and the risk of cancer in diverse populations.

Another important aspect of our analysis was to assess the potential association of -765G/C and -1195G/A COX-2 gene polymorphisms with clinical data of patients with PC. In the current study, the homozygous -1195AA was found to be present more frequently in patients with larger tumor size. To the best of our knowledge there are no available data about relationships between -1195G/A COX-2 polymorphism and clinical characteristics of patients with PC. Earlier, Tan et al[20] demonstrated that the COX-2 -1195A allele was associated with the presence of distant metastases in patients with colorectal cancer. They suggested that COX-2 may play a role not only in colorectal tumorigenesis but also in cancer progression by stimulating cell proliferation and spread.

According to our data, the second analyzed COX-2 polymorphism, -765G/C, was not associated with clinical parameters. Similarly, in other studies the -765G/C variant was not correlated with clinical stage of patients with cervical and colorectal cancers[17,21].

Overexpression of COX-2 may be an important cellular mechanism in smoking-related PC development. Numerous studies have shown that smoking induces COX-2 expression, but the exact signal pathways remain to be elucidated. Zhao et al[8] suggested that COX-2 genetic polymorphisms may determine interindividual variation in the inducibility of COX-2 expression. They observed that smoking remarkably increased COX-2 promoter activity, especially in patients with PC carrying the -765C allele. In contrast, in our study, there was no association between analyzed polymorphisms and smoking. Similarly, Pandey et al[21] did not find a correlation between smoking and COX-2 polymorphisms in patients with cervical cancer. This lack of association could be due to a relatively small number of subjects and certainly needs further validation.

In summary, we found a significant difference in the -1195G/A COX-2 gene polymorphism distribution between patients with PC and the control group. The presence of -1195AA genotype was associated with an increased PC risk; however, further studies are needed to investigate its possible association with PC prognosis. Our results are consistent with the biological function of the polymorphisms and support the hypothesis that aberrant arachidonic acid metabolism may play an important role in pancreatic carcinogenesis.

COMMENTS
Background

Despite improved diagnostic and therapeutic capabilities, pancreatic cancer (PC) still has a very poor prognosis. Numerous studies suggest a role for cyclooxygenase-2 (COX-2) in pancreatic carcinogenesis. COX-2 is involved in many biologic processes, such as cell proliferation, invasion, angiogenesis and inhibition of apoptosis, which are all relevant to cancer development and progression.

Research frontiers

The COX-2 gene was demonstrated to be genetically polymorphic, which may affect the expression or activity of this enzyme and consequently contribute to variation in individual susceptibility and aggressiveness of PC.

Innovations and breakthroughs

This study analyzed the clinical significance of -765G/C and -1195G/A COX-2 gene polymorphisms in patients with PC. In the study, the presence of the -1195AA genotype was associated with an increased risk of PC; however, further studies are needed to investigate its possible association with PC prognosis.

Applications

The results are consistent with the biological function of the polymorphisms and support the hypothesis that aberrant arachidonic acid metabolism may play an important role in pancreatic carcinogenesis.

Terminology

COX-2 is a key enzyme in the arachidonic acid pathway, initiating the synthesis of biologically important prostaglandin H2 (the precursor of other prostaglandins), prostacyclin and tromboxanes.

Peer review

In this manuscript, the authors demonstrate that COX-2 gene polymorphisms might be associated with carcinogenesis of PC. A series of experiments are well-planned and well-performed and this manuscript is well written.

Footnotes

Peer reviewer: Kotaro Miyake, MD, PhD, Department of Surgery, Institute of Health Biosciences, The University of Tokushima Graduate School, 3-18-15 Kuramoto, Tokushima 770-8503, Japan

S- Editor Tian L L- Editor Logan S E- Editor Xiong L

References
1.  Sobolewski C, Cerella C, Dicato M, Ghibelli L, Diederich M. The role of cyclooxygenase-2 in cell proliferation and cell death in human malignancies. Int J Cell Biol. 2010;2010:215158.  [PubMed]  [DOI]
2.  Ghosh N, Chaki R, Mandal V, Mandal SC. COX-2 as a target for cancer chemotherapy. Pharmacol Rep. 2010;62:233-244.  [PubMed]  [DOI]
3.  Juuti A, Louhimo J, Nordling S, Ristimäki A, Haglund C. Cyclooxygenase-2 expression correlates with poor prognosis in pancreatic cancer. J Clin Pathol. 2006;59:382-386.  [PubMed]  [DOI]
4.  Hermanova M, Karasek P, Tomasek J, Lenz J, Jarkovsky J, Dite P. Comparative analysis of clinicopathological correlations of cyclooxygenase-2 expression in resectable pancreatic cancer. World J Gastroenterol. 2010;16:1879-1884.  [PubMed]  [DOI]
5.  Bergmann F, Moldenhauer G, Herpel E, Gaida MM, Strobel O, Werner J, Esposito I, Müerköster SS, Schirmacher P, Kern MA. Expression of L1CAM, COX-2, EGFR, c-KIT and Her2/neu in anaplastic pancreatic cancer: putative therapeutic targets? Histopathology. 2010;56:440-448.  [PubMed]  [DOI]
6.  Abe T, Fukushima N, Brune K, Boehm C, Sato N, Matsubayashi H, Canto M, Petersen GM, Hruban RH, Goggins M. Genome-wide allelotypes of familial pancreatic adenocarcinomas and familial and sporadic intraductal papillary mucinous neoplasms. Clin Cancer Res. 2007;13:6019-6025.  [PubMed]  [DOI]
7.  Matsumoto G, Muta M, Tsuruta K, Horiguchi S, Karasawa K, Okamoto A. Tumor size significantly correlates with postoperative liver metastases and COX-2 expression in patients with resectable pancreatic cancer. Pancreatology. 2007;7:167-173.  [PubMed]  [DOI]
8.  Zhao D, Xu D, Zhang X, Wang L, Tan W, Guo Y, Yu D, Li H, Zhao P, Lin D. Interaction of cyclooxygenase-2 variants and smoking in pancreatic cancer: a possible role of nucleophosmin. Gastroenterology. 2009;136:1659-1668.  [PubMed]  [DOI]
9.  Papafili A, Hill MR, Brull DJ, McAnulty RJ, Marshall RP, Humphries SE, Laurent GJ. Common promoter variant in cyclooxygenase-2 represses gene expression: evidence of role in acute-phase inflammatory response. Arterioscler Thromb Vasc Biol. 2002;22:1631-1636.  [PubMed]  [DOI]
10.  Zhang X, Miao X, Tan W, Ning B, Liu Z, Hong Y, Song W, Guo Y, Zhang X, Shen Y. Identification of functional genetic variants in cyclooxygenase-2 and their association with risk of esophageal cancer. Gastroenterology. 2005;129:565-576.  [PubMed]  [DOI]
11.  Lurje G, Nagashima F, Zhang W, Yang D, Chang HM, Gordon MA, El-Khoueiry A, Husain H, Wilson PM, Ladner RD. Polymorphisms in cyclooxygenase-2 and epidermal growth factor receptor are associated with progression-free survival independent of K-ras in metastatic colorectal cancer patients treated with single-agent cetuximab. Clin Cancer Res. 2008;14:7884-7895.  [PubMed]  [DOI]
12.  Chiang SL, Chen PH, Lee CH, Ko AM, Lee KW, Lin YC, Ho PS, Tu HP, Wu DC, Shieh TY. Up-regulation of inflammatory signalings by areca nut extract and role of cyclooxygenase-2 -1195G&amp; gt; a polymorphism reveal risk of oral cancer. Cancer Res. 2008;68:8489-8498.  [PubMed]  [DOI]
13.  Bi N, Yang M, Zhang L, Chen X, Ji W, Ou G, Lin D, Wang L. Cyclooxygenase-2 genetic variants are associated with survival in unresectable locally advanced non-small cell lung cancer. Clin Cancer Res. 2010;16:2383-2390.  [PubMed]  [DOI]
14.  Coskunpinar E, Eraltan IY, Turna A, Agachan B. Cyclooxygenase-2 gene and lung carcinoma risk. Med Oncol. 2010;In press.  [PubMed]  [DOI]
15.  Kristinsson JO, van Westerveld P, te Morsche RH, Roelofs HM, Wobbes T, Witteman BJ, Tan AC, van Oijen MG, Jansen JB, Peters WH. Cyclooxygenase-2 polymorphisms and the risk of esophageal adeno- or squamous cell carcinoma. World J Gastroenterol. 2009;15:3493-3497.  [PubMed]  [DOI]
16.  Pereira C, Pimentel-Nunes P, Brandão C, Moreira-Dias L, Medeiros R, Dinis-Ribeiro M. COX-2 polymorphisms and colorectal cancer risk: a strategy for chemoprevention. Eur J Gastroenterol Hepatol. 2010;22:607-613.  [PubMed]  [DOI]
17.  Hoff JH, te Morsche RH, Roelofs HM, van der Logt EM, Nagengast FM, Peters WH. COX-2 polymorphisms -765G--&amp; gt; C and -1195A--&amp; gt; G and colorectal cancer risk. World J Gastroenterol. 2009;15:4561-4565.  [PubMed]  [DOI]
18.  Gangwar R, Mandhani A, Mittal RD. Functional polymorphisms of cyclooxygenase-2 (COX-2) gene and risk for urinary bladder cancer in North India. Surgery. 2011;149:126-134.  [PubMed]  [DOI]
19.  Dong J, Dai J, Zhang M, Hu Z, Shen H. Potentially functional COX-2-1195G&amp; gt; A polymorphism increases the risk of digestive system cancers: a meta-analysis. J Gastroenterol Hepatol. 2010;25:1042-1050.  [PubMed]  [DOI]
20.  Tan W, Wu J, Zhang X, Guo Y, Liu J, Sun T, Zhang B, Zhao D, Yang M, Yu D. Associations of functional polymorphisms in cyclooxygenase-2 and platelet 12-lipoxygenase with risk of occurrence and advanced disease status of colorectal cancer. Carcinogenesis. 2007;28:1197-1201.  [PubMed]  [DOI]
21.  Pandey S, Mittal RD, Srivastava M, Srivastava K, Mittal B. Cyclooxygenase-2 gene polymorphisms and risk of cervical cancer in a North Indian population. Int J Gynecol Cancer. 2010;20:625-630.  [PubMed]  [DOI]
22.  Talar-Wojnarowska R, Gasiorowska A, Olakowski M, Lekstan A, Lampe P, Malecka-Panas E. Clinical value of serum neopterin, tissue polypeptide-specific antigen and CA19-9 levels in differential diagnosis between pancreatic cancer and chronic pancreatitis. Pancreatology. 2010;10:689-694.  [PubMed]  [DOI]