Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Prognostic Role of Host Cyclooxygenase and Cytokine Genotypes in a Caucasian Cohort of Patients with Gastric Adenocarcinoma

  • María Asunción García-González ,

    asgarcia@unizar.es

    Affiliations Instituto de Investigación Sanitaria Aragón (IIS Aragón), Zaragoza, Spain, Centro de Investigación Biomédica en Red en enfermedades hepáticas y digestivas (CIBERehd), Barcelona, Spain

  • David Nicolás-Pérez,

    Affiliation Department of Gastroenterology, Hospital Universitario de Canarias, Tenerife, Spain

  • Angel Lanas,

    Affiliations Instituto de Investigación Sanitaria Aragón (IIS Aragón), Zaragoza, Spain, Centro de Investigación Biomédica en Red en enfermedades hepáticas y digestivas (CIBERehd), Barcelona, Spain, Faculty of Medicine, University of Zaragoza, Zaragoza, Spain, Department of Gastroenterology, Hospital Clínico Universitario, Zaragoza, Spain

  • Luis Bujanda,

    Affiliations Centro de Investigación Biomédica en Red en enfermedades hepáticas y digestivas (CIBERehd), Barcelona, Spain, Department of Gastroenterology, Hospital Donostia, Faculty of Medicine, University of Basque Country, San Sebastián, Spain

  • Patricia Carrera,

    Affiliation Centro de Investigación Biomédica en Red en enfermedades hepáticas y digestivas (CIBERehd), Barcelona, Spain

  • Rafael Benito,

    Affiliations Centro de Investigación Biomédica en Red en enfermedades hepáticas y digestivas (CIBERehd), Barcelona, Spain, Faculty of Medicine, University of Zaragoza, Zaragoza, Spain, Department of Microbiology, Hospital Clínico Universitario Lozano Blesa, Zaragoza, Spain

  • Mark Strunk,

    Affiliations Instituto de Investigación Sanitaria Aragón (IIS Aragón), Zaragoza, Spain, Centro de Investigación Biomédica en Red en enfermedades hepáticas y digestivas (CIBERehd), Barcelona, Spain

  • Federico Sopeña,

    Affiliations Instituto de Investigación Sanitaria Aragón (IIS Aragón), Zaragoza, Spain, Centro de Investigación Biomédica en Red en enfermedades hepáticas y digestivas (CIBERehd), Barcelona, Spain, Department of Gastroenterology, Hospital Clínico Universitario, Zaragoza, Spain

  • Santos Santolaria,

    Affiliation Department of Gastroenterology, Hospital San Jorge, Huesca, Spain

  • Elena Piazuelo,

    Affiliations Instituto de Investigación Sanitaria Aragón (IIS Aragón), Zaragoza, Spain, Centro de Investigación Biomédica en Red en enfermedades hepáticas y digestivas (CIBERehd), Barcelona, Spain

  • Pilar Jiménez,

    Affiliation Centro de Investigación Biomédica en Red en enfermedades hepáticas y digestivas (CIBERehd), Barcelona, Spain

  • Rafael Campo,

    Affiliation Department of Gastroenterology, Hospital Parc Tauli, Sabadell, Spain

  • Jesús Espinel,

    Affiliation Department of Gastroenterology, Complejo Hospitalario de León, León, Spain

  • Marisa Manzano,

    Affiliation Department of Gastroenterology, Hospital 12 de Octubre, Madrid, Spain

  • Fernando Geijo,

    Affiliation Department of Gastroenterology, Hospital Clínico Universitario, Salamanca, Spain

  • María Pellisé,

    Affiliation Department of Gastroenterology, Hospital Clinic i Provincial, Barcelona, Spain

  • Ferrán González-Huix,

    Affiliation Department of Gastroenterology, Hospital Josep Trueta, Girona, Spain

  • Jorge Espinós,

    Affiliation Department of Gastroenterology, Hospital Mútua de Terrassa, Spain

  • Manuel Zaballa,

    Affiliation Department of Gastroenterology, Hospital de Cruces, Baracaldo, Spain

  • Llúcia Titó,

    Affiliation Department of Gastroenterology, Hospital de Mataró, Mataró, Spain

  • Luis Barranco,

    Affiliation Department of Gastroenterology, Hospital del Mar, Barcelona, Spain

  • Roberto Pazo,

    Affiliation Department of Oncology, Hospital Miguel Servet, Zaragoza, Spain

  •  [ ... ],
  • Enrique Quintero

    Affiliation Department of Gastroenterology, Hospital Universitario de Canarias, Tenerife, Spain

  • [ view all ]
  • [ view less ]

Abstract

Background

Genetic factors influencing the prognosis of gastric adenocarcinoma (GAC) are not well known. Given the relevance of cytokines and other pro-inflammatory mediators in cancer progression and invasiveness, we aimed to assess the prognostic role of several functional cytokine and cyclooxygenase gene polymorphisms in patients with GAC.

Methodology

Genomic DNA from 380 Spanish Caucasian patients with primary GAC was genotyped for 23 polymorphisms in pro-inflammatory (IL1B, TNFA, LTA, IL6, IL12p40), anti-inflammatory (IL4, IL1RN, IL10, TGFB1) cytokine, and cyclooxygenase (PTGS1 and PTGS2) genes by PCR, RFLP and TaqMan assays. Clinical and histological information was collected prospectively. Survival curves were estimated by the Kaplan-Meier method and compared using the log rank test. Outcome was determined by analysis of Cox proportional hazards, adjusting for confounding factors.

Results

The median follow-up period and median overall survival (OS) time were 9.9 months (range 0.4–120.3) and 10.9 months (95% CI: 8.9–14.1), respectively. Multivariate analysis identified tumor stages III (HR, 3.23; 95% CI:2–5.22) and IV (HR, 5.5; 95% CI: 3.51–8.63) as independent factors associated with a significantly reduced OS, whereas surgical treatment (HR: 0.44; 95%CI: 0.3–0.6) was related to a better prognosis of the disease. Concerning genetic factors, none of the 23 polymorphisms evaluated in the current study did influence survival. Moreover, no gene-environment interactions on GAC prognosis were observed.

Conclusions

Our results show that, in our population, the panel of selected pro- and anti-inflammatory cytokine, and cyclooxygenase gene polymorphisms are not relevant in determining the prognosis of gastric adenocarcinoma.

Introduction

Gastric adenocarcinoma (GAC) still remains the second leading cause of cancer death worldwide. Despite advances in treatment, overall 5-year survival is less than 40% in Western countries [1], [2]. Poor prognosis of patients with GAC has been associated with several conditions such as tumor-node-metastasis (TNM) stage [3], signet-ring cell histology [4], incomplete tumor resection [5], and high lymph node ratio [6]. Helicobacter pylori (H. pylori) infection and tobacco smoking, two well establish risk factors for gastric carcinogenesis [7][9] have been also evaluated as prognostic markers for GAC [10][13]. However, few studies have addressed the relevance of patient’s genetic background on the prognosis of the disease. Recently, it has been suggested the correlation between persistence of chronic inflammation and reduced survival in GAC patients [14], [15]. The immune response at the gastrointestinal mucosa is regulated by a wide variety of pro- and anti-inflammatory mediators such as cytokines and growth factors. Among them, interleukin-1β (IL-1β), tumor necrosis factor-α (TNF-α), lymphotoxin-α (LT-α), interleukin-6 (IL-6), and interleukin-12 (IL-12) are potent pro-inflammatory cytokines with a relevant role in both cancer development and progression. Significant high serum levels of IL-1β, TNF-α, and IL-6 have been reported in patients with metastatic GAC [16], [17]. Moreover, upregulation of IL-1β and TNF-α seems to be involved in the molecular mechanisms of the anorexia-cachexia syndrom [18]. By contrast, IL-12 production by PBMCs (peripheral blood mononuclear cells) after stimulation in vitro with tumor cells or LPS (lipopolysaccharide) has been found to be significantly depressed in patients with GAC, specially in the advanced stages of the disease [19].

On the other hand, IL-4, IL-10, and TGF-β1 are potent negative regulators of the Th1-type immune response. Studies carried out in human gastric cancer cell lines found that IL-4 inhibits gastric cell proliferation by blocking cell cycle progression and down-regulating several key G0–G1 cell cycle nuclear factors [20]. IL-10 inhibits macrophage activation, cytokine production, and antigen-specific T-cell proliferation [21]. However, several studies have postulated the involvement of IL-10 in the onset and spread of GAC [22]. Like IL-10, enhanced expression of TGF-β1 has been associated with progression and invasiveness of the GAC [23], [24]. Thus, blockade of TGF-β or TGF-β signaling pathways has been suggested as potential therapy to prevent GAC cells from invading and metastasizing [25].

Besides cytokines and growth factors, cyclooxygenases 1 (COX-1) and 2 (COX-2) have been implicated in carcinogenesis and metastatic progression of many types of cancers including GAC [26], [27]. Cyclooxigenases or prostaglandin G/H synthases (PTGSs) catalyze the formation of prostaglandins from arachidonic acid [28]. COX-1 (PTGS-1) is constitutively expressed and plays a key role in the protection of gastric mucosa. The COX-2 isoform (PTGS-2) is induced in response to cytokines, and other inflammatory and mitogenic stimuli. Whereas the mechanisms by which COX-1 promotes gastric carcinogenesis are not well known, up-regulation of COX-2 has been associated with GAC development and progression by increasing cell proliferation, inhibiting apoptosis, inducing angiogenesis, and suppressing host immune response [29].

Genes encoding the proteins mentioned above harbor polymorphic sites that have been reported to influence transcriptional efficiency and protein levels. At present, there is very little information concerning the influence of immune-related gene variants on the prognosis and clinical outcome of GAC [14], [30][32]. Trying to address this specific issue, we aimed to assess the prognostic role of some functional polymorphisms in pro- (IL1B, TNFA, LTA, IL6, IL12p40), and anti-inflammatory (IL1RN, IL4, IL10, TGFB1) cytokine as well as in the PTGS1 and PTGS2 genes in a large prospective cohort of Spanish Caucasian patients with primary GAC.

Patients and Methods

Study Subjects and Data Collection

Consecutive patients diagnosed with primary GAC from May 2002 to December 2003 in 15 general Spanish hospitals were invited to take part in the study. Gastric tumors were grouped according to their anatomical location as cardia GAC (located at the gastroesophageal junction) [33] and non-cardia CGA. Moreover, non-cardia GACs were classified according to the histological type as intestinal, diffuse, or indeterminate [34]. Patients with local recurrence of GAC, non-adenocarcinoma histology, absence of blood samples, or refusal to participate in the study were considered non-eligible. Of the 466 patients with GAC who initially agreed to participate, 400 (85.1%) could be interviewed, had a complete pathology report, and provided biological samples of adequate quality for genetic analysis. However, 20 patients (5%) were excluded from the study due to lack of follow-up. Finally, 380 GAC patients had adequate information to estimate follow-up data and survival analysis. The characteristics of excluded patients did not differ from those of the final group.

At the time of inclusion, detailed information was recorded concerning age, gender, smoking habits, family history of GAC, date of diagnosis, surgical treatment, TNM staging (UICC/AJCC classification), presence of metastases, tumor location, and histological subtype. In addition, approximately 10 ml of peripheral blood from each patient were collected into ethylenediaminetetraacetic acid (EDTA) and serum separator tubes for subsequent DNA extraction and H. pylori serology, respectively. Once processed, whole blood and serum samples were aliquoted and stored at –80°C until analysis. All patients gave written informed consent to the study, which was reviewed and approved by the institutional ethics committee of each participating hospital.

Patient’s Follow-up

Each participant hospital performed the follow-up periodically. Follow-up included computerized tomography of the chest and abdomen, and hematological analysis at 3-month intervals during the first year and thereafter at 6-month intervals. Moreover, an upper digestive endoscopy was performed every year. Information was updated by clinical specialists through in-person interview, medical chart review, and in some cases direct calling. The latest follow-up data in this study were obtained in November 2011.

Diagnosis of Helicobacter pylori Infection

H. pylori status was assessed by both urease test (CLO-test; Delta West Ltd., Canning Vale, Bentley, Australia) and histological examination from biopsies taken at the antrum and corpus of the stomach during the endoscopic procedure. The presence of antibodies to CagA and VacA antigens was determined in serum by Western blot analysis (Bioblot Helicobacter; Izasa, Barcelona, Spain), as previously validated in our area [35]. Patients were considered positive for bacterial infection if any of the three tests was positive.

Genetic Polymorphisms

Genomic DNA was extracted from EDTA-preserved whole blood using the QIAamp DNA Blood Mini extraction kit (Qiagen, Izasa, Barcelona, Spain). We assessed 22 SNP’s (single nucleotide polymorphisms) in the IL1B, TNFA, LTA, IL12p40, IL4, IL6, IL10, TGFB1, PTGS1, and PTGS2 genes, as well as the VNTR (variable number of tandem repeat) polymorphism in intron 2 of the IL1RN gene. The panel of polymorphisms was selected a priori based on three criteria: (a) having a reported prevalence of at least 5% for the less frequent allele among Caucasians; (b) having potential functional consequences leading to altered protein concentrations or protein function; or (c) published evidence of their involvement in GAC development, progression and invasiveness. The regions containing the polymorphic sites were blindly genotyped by RFLP (restriction fragment length polymorphism)-PCR-based methods and TaqMan®-MGB allelic discrimination assays. Quality control for the genotyping was achieved by including a negative PCR control sample (HPLC water) and three positive controls for each SNP analyzed (homozygous for allele 1, heterozygous, and homozygous for allele 2). In addition, 10% of the samples were run twice in separate assays with a genotype concordance of 100% for all the polymorphisms.

Cytokine gene analysis.

Thirteen SNP’s in the IL1B (−511C>T, rs16944 and +3954C>T, rs1143634), TNFA (−308G>A, rs1800629 and −238G>A, rs361525), LTA (+252G>A, rs909253 and +365G>C, rs746868), IL12p40 (+1180A>C, rs3212227), IL4 (−590C>T, rs2243250), IL6 (+174C>G, rs1800795), IL10 (−1087G>A, rs1800896 and −597C>A, rs2243250), and TGFB1 (+869T>C, rs1800470 and +915G>C, rs1800471) genes, as well as the VNTR polymorphism in intron 2 of the IL-1RN gene were analysed by PCR-RFLP-based methods and TaqMan®-MGB assays (Applied Biosystems, Madrid, Spain) as previously described [36].

PTGS1 genotyping.

Subjects were genotyped for three SNPs in the PTGS1 gene. The PTGS1-1676A>G (rs1330344) and +644C>A (rs5788) polymorphisms were analyzed using TaqMan®-MGB pre-designed assays according to the manufacturer’s instructions. The fragment containing the PTGS1+50C>T (Pro17Leu) (rs3842787) polymorphic site was amplified as described by Gonzalez-Conejero et al. [37]. PCR products were digested with 2 units of Fau I (New England Biolabs, Izasa, Barcelona, Spain) for 5 hours at 55°C and electrophoresed on 2% agarose gels. Digests resulted in an intact fragment of 244 bp (allele T) or in two fragments of 125 and 119 bp (allele C).

PTGS2 genotyping.

Six SNPs located within the PTGS2 gene were studied. TaqMan® Pre-Designed Assays were used for the detection of the PTGS2 -1195G>A, (rs689466), +3050 (V102V)G>C, (rs5277), +8473T>C, (rs5275) and +10335G>A, (rs689469) polymorphisms. Genotypes for the PTGS2-765G>C (rs20417) and +9850A>G (rs4648298) polymorphisms were determined by PCR-RFLP based methods using primers and reaction profiles as described by Cipollone et al. [38] and Cox et al. [39], respectively. For typing the rs20417 polymorphism, PCR products were digested with 2 units of Fau I (New England Biolabs, Izasa, Barcelona, Spain) for 5 hours at 55°C; digests resulted in two fragments of 122 and 187 bp (allele C) or in an intact fragment of 309 bp (allele G). For typing the rs4648298 polymorphism, PCR products were digested with 2 U of Alu I (Invitrogen, Prat de Llobregat, Barcelona, Spain) for 5 h at 37°C. This procedure resulted in 349-bp and 196-bp fragments (allele G) or in the undigested 545-bp fragment (allele A).

Statistical Analysis

Continuous variables were expressed as mean with standard deviation whereas qualitative variables were expressed as frequencies and percentages. The relationship between qualitative variables was analyzed by contingency tables with chi-square test (χ2). Overall survival (OS) time was calculated from the date of the diagnosis to the date of last contact or death from any cause. Patients who where still alive at the last contact and patients lost to follow-up were consider as a censored event in the analysis. In addition, patients’ comorbidity at diagnosis was assessed using a previously validated adaptation of the Charlson Comorbidity Index [40]. Concerning gene polymorphisms, estimated haplotype frequencies and linkage disequilibrium (LD) coefficients (D´and r2) for the TGFB1, TNFA, LTA, IL10, IL1, PTGS1, and PTGS2 loci were calculated using the Estimating Haplotype frequencies (EH) software program (available from http://linkage.rockefeller.edu/ott/eh.htm. For each marker, the more common homozygous genotype or haplotype was used as the reference category. Co-dominant and dominant inheritance genetic models were used for analysis. Survival among different genotype groups was estimated using the Kaplan-Meier method and compared using the log rank test. Univariate and multivariate Cox proportional hazards models adjusting for age, gender, H. pylori status, smoking habits, tumor location, histological type, tumor stage, treatment, and Charlson index, were performed to evaluate the prognostic value of each polymorphism on patient’s survival. Starting with age and sex, models were constructed using a step-wise forward unconditional method. A variable was entered in the model if the significance level of its coefficient was less than 0.05 and was removed if it was greater than 0.10. Potential interactions between genotypes and clinical and demographic variables were assessed by the corresponding Cox regression models containing the interaction term. For all tests, a two-sided p-value <0.05 was considered statistically significant. To address the issue of conducting multiple tests within each polymorphism, an additional Bonferroni correction was applied (P-value <0.05/46 = P-value <0.001). The statistical analyses were performed using the SPSS software v 15.0 for Windows (SPSS Ibérica, Madrid, Spain).

Results

Survival Analysis and Characteristics of GAC Patients

Demographic, clinical, and tumor-related characteristics of patients included in the study are summarized in Table 1. There were 257 males (67.6%) and 123 females (32.4%, male/female ratio 2/1) whose ages ranged from 30 to 96 years. According to Charlson’s index, most of patients (87.6%) had a low morbidity index (<3) at the moment of diagnosis. In 63 cases (16.6%), tumors were located at the cardia, and in 317 cases (83.4%), at the distal region of the estomach. Among distal GACs, 161 (50.8%) were of intestinal histotype, 119 (37.5%) of diffuse histotype, and 37 (11.7%) of mixed or undetermined type.

thumbnail
Table 1. Demographic and clinicopathological characteristics of patients with GAC (n = 380).

https://doi.org/10.1371/journal.pone.0046179.t001

The median follow-up time and the median OS for all patients in our study was 9.9 months (range: 0.4–120.3) and 10.9 months (CI 95%: 8.9–14.1) respectively. The mean and median follow-up for censored patients were 67±42 and 73.4 months (range: 0.6–116.1), respectively. Three hundred and eleven GAC patients (81.8%) had died at the end of the follow-up period, and in 227 cases (73%) death was related to tumor progression. Figure 1 shows Kaplan-Meier survival curves among GAC patients regarding clinicopathological features. In the univariate analysis, male gender, previous or current smoking, proximal location of the tumor, and advanced tumor stages (III and IV) were associated with significantly reduced OS, whereas surgical treatment and a low morbidity index (<3) at the moment of diagnosis were related to a better prognosis of the disease (Table 2). By contrast, other clinicopathological features evaluated in our study such as age, tumor histological type, H. pylori status, or treatment with D1 or D2 lymphadenectomy did not influence survival.

thumbnail
Figure 1. Kaplan-Meier survival plots presented by (A) TNM stage, (B) surgical treatment, (C) age, (D) gender, (E) smoking habit, (F) H. pylori infection, (G) tumor location, and (H) histological subtype.

Statistical analysis was performed by the log rank test.

https://doi.org/10.1371/journal.pone.0046179.g001

thumbnail
Table 2. Overall survival analysis according to clinicopathological features.

https://doi.org/10.1371/journal.pone.0046179.t002

Survival Analysis and Gene Polymorphisms

Tables 3 and 4 show the survival hazards ratios among GAC patients according to gene polymorphisms. In the univariate analysis, no significant differences were found when GAC patients were analyzed by genotypes for any of the polymorphisms studied. Only some borderline associations between the TNFA rs1800629 (p = 0.07 for the AA genotype), LTA rs909253 (p = 0.06 for the AG genotype), IL10 rs2243250 (p = 0.04 for the CA genotype), and PTGS2 rs4648298 (p = 0.06 for the AG genotype) and OS was observed. Because there were few GAC patients homozygous for the minor alleles of the different polymorphisms, the heterozygous and minor variant homozygous genotypes were combined for additional analysis, assuming a dominant genetic model. Again, no association between OS and any polymorphisms was observed.

thumbnail
Table 3. Overall survival analysis according to cytokine gene polymorphisms.

https://doi.org/10.1371/journal.pone.0046179.t003

thumbnail
Table 4. Overall survival analysis according to PTGS gene polymorphisms.

https://doi.org/10.1371/journal.pone.0046179.t004

Further stratification of patients (Tables S1, S2, and S3 and Figures S1, S2, S3, and S4) by tumor location (cardia vs. non-cardia) and histological subtype (intestinal vs. diffuse) showed no differences in OS among the different gene variants evaluated in the study (Table S4).

An additional haplotype analysis was conducted to evaluate the combined effect of alleles of different polymorphisms on gastric cancer survival. As expected, the IL1, TNFA, LTA, TGFB1, IL10, PTGS1, and PTGS2 loci were in strong linkage disequilibrium (LD) in our data set (data not shown). Specific LD values for each pair of loci of IL1, TNFA, LTA, TGFB1, and IL10 genes were similar to those reported previously [36]. PTGS1 and PTGS2 polymorphisms were also in strong LD with D' values higher than 0.75 with the exception of the PTGS1-1676/PTGS1 644 loci (D' = 0.21). Table 5 shows the OS analysis according to the most frequent cytokine and PTGS haplotypes. As with the independent analysis for each polymorphism, none of the estimated haplotypes showed a significant association with OS.

thumbnail
Table 5. Overall survival analysis according to cytokine and PTGS estimated haplotypes.

https://doi.org/10.1371/journal.pone.0046179.t005

Despite the lack of influence of polymorphisms on patient’s survival, we wanted to examine the potential interaction between genotypes and the prognostic factors of GAC identified in our study population (namely TNM stage and surgical treatment). Smoking habit, and H. pylori infection, two well known risk factors of GCA were also included for gene-interaction analysis. The exposure variables were as follows: TNM stage (was codified as a continuous variable), surgical treatment (treated vs. untreated), smoking habit (never smokers vs. current smokers), and H. pylori infection (positive vs. negative). Tests for interaction under a multiplicative model showed no significant association of any SNPs with tumor stage, surgery, smoking, and H. pylori status in relation to survival (Tables S5 and S6). Only a weak significant interaction between the PTGS2 rs4648298 AA genotype with advance stages and reduced OS was observed (P interaction = 0.036).

Gene-gene interactions between all cytokine and PTGS polymorphisms were also investigated (Tables S7 and S8). Among all interactions evaluated, the specific interaction between PTGS1 rs5788 and TGFB1 rs1800470 gene polymorphisms reached the highest value. Thus, GAC patients carrying both mutant alleles (rs5788A/rs1800470C) had a better OS than non-carriers (median survival 595 days vs. 309; P-value = 0.007 by the log rank test; Pinteraction = 0.002). However, after correction for multiple testing, the association did not reach statistical significance (an interaction term P-value <0.00001 was considered statistically significant after correction for the number of interactions tested [23×23 = 529; 0.05/529 = 9.5 10−5].

In summary, of the environmental and clinicopathologic features evaluated in this study and after controlling for confounding factors, Cox regression analysis identified tumor stages III (HR, 3.23; 95% CI: 2–5.22) and IV (HR, 5.5; 95% CI: 3.51–8.63) as independent factors associated with significantly reduced OS, whereas surgical treatment (HR: 0.44;95%CI: 0.3–0.6) was related to a better prognosis of the disease (Table 6). Concerning genetic factors, none of the polymorphisms evaluated in the current study were related to GAC prognosis. Moreover, no interactions between any SNP’s and the identified prognostic factors were observed.

thumbnail
Table 6. Multivariate Cox proportional hazard analysis for GAC patients.

https://doi.org/10.1371/journal.pone.0046179.t006

Discussion

Since the publication in 2000 of the first landmark report by El-Omar and co-workers [41] reporting the association of IL-1B rs1143627T and IL1RN*2/*2 genotypes with an increased risk of GAC, numerous studies concerning the association of cytokine gene polymorphisms and GAC risk have been conducted. However, few studies have addressed the role of cytokine gene and other immune mediators on the prognosis of the disease. We analyzed in this study a total of 23 polymorphisms localized in 11 immune-related genes (IL1B, TNFA, LTA, IL6, IL12p40, IL1RN, IL4, IL10, TGFB1, PTGS1, and PTGS2). Concerning cytokine genes, none of the polymorphisms analyzed in the current study were related to GAC prognosis. This finding is in contrast with those reported previously in two European studies conducted by Graziano et al. [30] and Deans et al. [14]. The former [30] reported the association of specific IL1B and IL1RN variants (IL1B rs16944T/IL1B rs1143627C and IL1RN 1) with shortened survival of patients with advanced GAC, and the later [14] showed the association of the pro-inflammatory haplotype IL6 rs1800795C/IL10 rs1800896G/TNFA rs1800629A with an adverse prognosis in patients with squamous esophageal carcinoma or gastroesophageal adenocarcinoma. However, the study by Graziano et al. [30] was carried out in a series of 123 GAC patients with recurrent or metastatic tumors treated with palliative chemotherapy, and the study by Deans et al. [14] comprised a mixed population of patients with GAC and patients with esophageal squamous cell carcinoma. Our study was performed in a large non-selected cohort of 380 Spanish Caucasian patients with primary GAC in which patients with secondary or recurrent tumors and patients with non-adenocarcinoma histology were excluded at entry. These differences in design and methodology, along with geographical variations in allele frequencies and the plausible effect of other SNPs in these genes could explain, at least in part, the discrepant results reported among studies.

With regard to cyclooxigenases, most studies have implicated COX-2 as the COX isoform involved in cancer development and invasiveness. By contrast, initial evidence for a role of COX-1 in carcinogenesis was scarce. Recent studies reported a high expression of COX-1 in malignant tissues of different types of cancer such as ovarian, breast, and esophageal cancers [42][44]. In addition, experimental studies in Min mice lacking the PTGS1 gene showed a reduction in intestinal polyp formation compared to the wild-type PTGS1+/Min mice [45]. Although the mechanisms by which COX-1 is implicated in carcinogenesis are not well known, it has been suggested that it is mediated through the induction of COX-2 expression via a paracrine mechanism [46], [47]. COX-1 and COX-2 proteins are encoded by the PTGS1 and PTGS2 genes which are known to be highly polymorphic. In the last years, several studies have addressed the role of PTGS1 and PTGS2 polymorphisms on GAC risk and its precursors with inconsistent results [48][53]. However, and to our knowledge, this is the first study evaluating the relevance of PTGS variants in the prognosis and survival of GAC patients. In our study none of the PTGS1 and PTGS2 polymorphisms were related to GAC prognosis. Only a weak interaction between the rs4648298 polymorphism in the PTGS2 gene with advance stages of the disease and reduced OS was observed. Recently, two studies by Iglesias et al [54] and Coghill et al [55] have reported the association of specific PTGS1 and PTGS2 gene variants with colorectal cancer survival. The former described a link between the PTGS2 rs4648298G variant and longer survival in a Spanish population of patients with colorectal cancer. The latter, identified four PTGS1 polymorphisms influencing colorectal cancer mortality. According to the authors, the rs12132666A variant was associated with 50% lower mortality whereas minor alleles of the rs10306155, rs4836885, and rs5789 resulted in significantly reduced patient’s survival. However, and as point out by the authors, the magnitude of these associations was attenuated after adjustment for stage of the disease at diagnosis. In any case, these studies highlight the need for further research to evaluate whether polymorphisms involved in the prostaglandin synthesis pathway may have the potential to predict survival in patients with GAC.

Similarly to what occurs to cytokine and PTGS genes, little information is available with regard to genes encoding growth factors. Among them, TGFB1 presents special interest due to its role in gastric cancer development, progression and invasiveness [23], [24]. Two functional polymorphisms at positions +869 T>C (rs1800470) and +915 G>C (rs1800471) in the signal protein sequence of the TGFB1 gene [56], [57] have been related to cancer progression and patient’s survival in several types of cancer [58], [59]. In the current study, neither the rs1800470 nor the rs1800471 polymorphisms were associated with overall survival in GAC patients. Moreover, no interaction with other clinicopathological features such as H. pylori status, smoking habit, TNM stage or surgical treatment was observed. In agreement with our results, Guan et al [60] found no association between rs1800470 and rs1800471 variants and OS rates, although in this case patients carrying the rs1800471C variant showed a poorer 2-year survival than non carriers.

Besides host factors, we also evaluate the relevance of environmental factors, namely H. pylori infection and tobacco smoking, as prognostic markers for GAC. We found that smoking habit was independently associated with a worse prognosis, especially in patients with earlier stages of the disease (HR: 1.78; 95% CI: 1.12–3.5). These findings are in agreement with one of the few studies addressing this issue in which habitual smoking was suggested as an adverse prognostic factor for gastric cancer in Japanese patients [13]. Nicotine is considered the major psychoactive compound of cigarette smoke and it has been well documented to play a key role in gastric cancer [61], [62]. However, its effect on angiogenesis and invasion remains largely unknown, although recent experimental studies have showed that nicotine can stimulate gastric cancer cell proliferation, migration and invasiveness through a COX-2/VEGF dependent pathway [63]. Regarding the influence of H. pylori status in GAC outcome, we found no association between bacterial infection and GAC prognosis. Recent studies have suggested that H. pylori infection may be related to better prognosis in patients with GAC. Meimarakis et al. [64] demonstrated that infection prior to curative-intent resection of GAC correlated with both higher relapse-free and OS rates in early stages cancers (T1 and T2). H. pylori status was also found to influence survival in patients with early as well as advanced stages of disease in a subsequent study by Marrelli et al. [10]. The reasons for this association are not known but it was explain on the basis of an improved immune response against the tumor induced by H. pylori [64]. However, this hypothesis was not confirmed by other studies [65], [66], including our own, suggesting that H. pylori negativity may be simply related with more advanced stages or progression of the disease.

Finally, our study has several strengths and limitations. This investigation was carried out in a homogeneous Caucasian population of Spanish patients with primary GAC followed for a long period of time. Moreover, and to our knowledge, this is the first study evaluating the relevance of PTGS1 and PTGS2 variants in the prognosis and survival of GAC patients. On the other hand, some limitations should be also considered. In particular, the relatively small sample size limited the power to detect smalls HRs in those low-frequency homozigous variant polymorphisms. Setting an α value of 0.05, the study had a power of 85% to detect HRs >1.4. As a result, it is possible that we could miss minor statistical differences especially when subgroup analyses and assessment of gene-environmental interactions were performed. A second limitation of the study was the lack of a centralized pathological assessment. Evaluation of biopsies and surgical specimens was accomplished at each participant hospital which may represent a source of bias since interobserver variability was not controlled. Variables affecting GAC survival in our study (TNM stage and surgical treatment) have been described as common prognostic factors in the Literature [14], [60], [67]. In the survival analysis, only surgical treatment information was considered for evaluation although data regarding chemo- and radiotherapy were also available. However, radiotherapy and chemotherapy schedules varied considerably among the participating hospitals which precluded a reliable assessment of their effects on disease outcome. Further studies evaluating potencial interactions between clinical and gene polymorphisms on GAC survival should take into consideration the relevance of including a detailed treatment information which could help to identify interactions that may have direct implications for therapy and follow-up strategies.

In summary, our data show that the specific polymorphisms among pro- and anti-inflammatory cytokine and PTGS gene polymorphisms evaluated in this study are not related to GAC prognosis in the Spanish population. However, we can not rule out that further studies with larger sample size could detect as statistically significant some small differences found in our study. Future genome-wide association studies (GWAS) and well designed studies in different areas and ethnic groups are needed in order to determine the real contribution of host genetic factors into the prognosis of gastric cancer.

Supporting Information

Figure S1.

Kaplan-Meier survival plots in cardia GAC patients. Kaplan-Meier survival plots in cardia GAC patients (n = 63) presented by (A) TNM stage, (B) surgical treatment, (C) age, (D) gender, (E) smoking habit, (F) and H. pylori infection status (G). Statistical analysis was performed by the log-rank test.

https://doi.org/10.1371/journal.pone.0046179.s001

(TIF)

Figure S2.

Kaplan-Meier survival plots in non-cardia GAC patients. Kaplan-Meier survival plots in non-cardia GAC patients (n = 317) presented by (A) TNM stage, (B) surgical treatment, (C) age, (D) gender, (E) smoking habit, (F) and H. pylori infection status. Statistical analysis was performed by the log-rank test.

https://doi.org/10.1371/journal.pone.0046179.s002

(TIF)

Figure S3.

Kaplan-Meier survival plots in intestinal GAC patients. Kaplan-Meier survival plots in intestinal GAC patients (n = 161) presented by (A) TNM stage, (B) surgical treatment, (C) age, (D) gender, (E) smoking habit, (F) and H. pylori infection status. Statistical analysis was performed by the log-rank test.

https://doi.org/10.1371/journal.pone.0046179.s003

(TIF)

Figure S4.

Kaplan-Meier survival plots in diffuse GAC patients. Kaplan-Meier survival plots in diffuse GAC patients (n = 119) presented by (A) TNM stage, (B) surgical treatment, (C) age, (D) gender, (E) smoking habit, (F) and H. pylori infection status. Statistical analysis was performed by the log-rank test.

https://doi.org/10.1371/journal.pone.0046179.s004

(TIF)

Table S1.

Demographic and clinicopathological characteristics of GAG patients stratified according to the location of the tumor (cardia/non-cardia) and histological subtype (intestinal/diffuse).

https://doi.org/10.1371/journal.pone.0046179.s005

(DOC)

Table S2.

Overall survival analysis and clinicopathological features in GAC patients stratified according to the location of the tumor (cardia/non-cardia).

https://doi.org/10.1371/journal.pone.0046179.s006

(DOC)

Table S3.

Overall survival analysis and clinicopathological features in GAC patients stratified according to the histological type of the tumor (intestinal/diffuse).

https://doi.org/10.1371/journal.pone.0046179.s007

(DOC)

Table S4.

Overall survival analysis and gene polymorphisms according to the location and histological subtype of the tumor.

https://doi.org/10.1371/journal.pone.0046179.s008

(DOC)

Table S5.

Interaction between cytokine gene polymorphisms and clinicopathological features.

https://doi.org/10.1371/journal.pone.0046179.s009

(DOC)

Table S6.

Interaction between PTGS gene polymorphisms and clinicopathological features.

https://doi.org/10.1371/journal.pone.0046179.s010

(DOC)

Table S7.

Gene-gene interactions between cytokine and PTGS gene polymorphisms.

https://doi.org/10.1371/journal.pone.0046179.s011

(DOC)

Table S8.

Gene-gene interactions between cytokine and PTGS gene polymorphisms.

https://doi.org/10.1371/journal.pone.0046179.s012

(DOC)

Acknowledgments

The authors thank Samantha Arechavaleta for her technical assistance in the genotyping analysis.

Author Contributions

Conceived and designed the experiments: MAG-G DN-P EQ L. Bujanda AL PC. Performed the experiments: RB MS EP PJ. Analyzed the data: PC FS SS RC J. Espinel MM FG MP FG-H J. Espinós MZ LT L. Barranco RP MAG-G. Contributed reagents/materials/analysis tools: MS RB. Wrote the paper: MAG-G DN-P.

References

  1. 1. Cunningham D, Allum WH, Stenning SP, Thompson JN, Van de Velde CJ, et al. (2006) Perioperative chemotherapy versus surgery alone for resectable gastroesophageal cancer. N Engl J Med. 355: 11–20.
  2. 2. Degiuli M, Sasako M, Ponti A, Calvo F (2004) Survival results of a multicentre phase II study to evaluate D2 gastrectomy for gastric cancer. Br J Cancer. 90: 1727–1932.
  3. 3. Kooby DA, Suriawinata A, Klimstra DS, Brennan MF, Karpeh MS (2003) Biologic predictors of survival in node-negative gastric cancer. Ann Surg. 237: 828–835.
  4. 4. Piessen G, Messager M, Leteurtre E, Jean-Pierre T, Mariette C (2009) Signet ring cell histology is an independent predictor of poor prognosis in gastric adenocarcinoma regardless of tumoral clinical presentation. Ann Surg. 250: 878–887.
  5. 5. Shiraishi N, Sato K, Yasuda K, Inomata M, Kitano S (2007) Multivariate prognostic study on large gastric cancer. J Surg Oncol. 96: 14–18.
  6. 6. Marchet A, Mocellin S, Ambrosi A, Morgagni P, Garcea D, et al. (2007) The ratio between metastatic and examined lymph nodes (N ratio) is an independent prognostic factor in gastric cancer regardless of the type of lymphadenectomy: results from an Italian multicentric study in 1853 patients. Ann Surg. 245: 543–552.
  7. 7. Uemura N, Okamoto S, Yamamoto S, Matsumura N, Yamaguchi S, et al. (2001) Helicobacter pylori infection and the development of gastric cancer. N Engl J Med 345: 784–789.
  8. 8. Helicobacter and Cancer Collaborative Group (2001) Gastric cancer and Helicobacter pylori: a combined analysis of 12 case control studies nested within prospective cohorts. Gut. 49: 347–353.
  9. 9. Ladeiras-Lopes R, Pereira AK, Nogueira A, Pinheiro-Torres T, Pinto I, et al. (2008) Smoking and gastric cancer: systematic review and meta-analysis of cohort studies. Cancer Causes Control. 19: 689–701.
  10. 10. Marrelli D, Pedrazzani C, Berardi A, Corso G, Neri A, et al. (2009) Negative Helicobacter pylori status is associated with poor prognosis in patients with gastric cancer. Cancer 115: 2071–2080.
  11. 11. Rad R, Prinz C, Schmid RM (2006) Helicobacter pylori and prognosis of gastric carcinoma. Lancet Oncol. 7: 364–365.
  12. 12. Kawaguchi T, Matsumura A, Fukai S, Tamura A, Saito R, et al. (2010) Japanese ethnicity compared with Caucasian ethnicity and never-smoking status are independent favorable prognostic factors for overall survival in non-small cell lung cancer: a collaborative epidemiologic study of the National Hospital Organization Study Group for Lung Cancer (NHSGLC) in Japan and a Southern California Regional Cancer Registry databases. J Thorac Oncol. 5: 1001–1010.
  13. 13. Huang XE, Tajima K, Hamajima N, Kodera Y, Yamamura Y, et al. (2000) Effects of dietary, drinking, and smoking habits on the prognosis of gastric cancer. Nutr Cancer. 38: 30–36.
  14. 14. Deans C, Rose-Zerilli M, Wigmore S, Ross J, Howell M, et al. (2007) Host cytokine genotype is related to adverse prognosis and systemic inflammation in gastro-oesophageal cancer. Ann Surg Oncol. 14: 329–339.
  15. 15. Rashid SA, O’Quigley J, Axon AT, Cooper EH (1982) Plasma protein profiles and prognosis in gastric cancer. Br J Cancer. 45: 390–394.
  16. 16. Kabir S, Daar GA (1995) Serum levels of interleukin-1, interleukin-6 and tumour necrosis factor-alpha in patients with gastric carcinoma. Cancer Lett 95: 207–212.
  17. 17. Ashizawa T, Okada R, Suzuki Y, Takagi M, Yamazaki T, et al. (2005) Clinical significance of interleukin-6 (IL-6) in the spread of gastric cancer: role of IL-6 as a prognostic factor. Gastric Cancer. 8: 124–131.
  18. 18. Inui A (2002) Cancer anorexia-cachexia syndrome: current issues in research and management. CA Cancer J Clin. 52: 72–91.
  19. 19. Shibata M, Nezu T, Kanou H, Abe H, Takekawa M, et al. (2002) Decreased production of interleukin-12 and type 2 immune responses are marked in cachectic patients with colorectal and gastric cancer. J Clin Gastroenterol 34: 416–420.
  20. 20. Morisaki T, Uchiyama A, Yuzuki D, Essner R, Morton DL, et al. (1994) Interleukin 4 regulates G1 cell cycle progression in gastric carcinoma cells. Cancer Res 54: 1113–1118.
  21. 21. de Waal Malefyt R, Haanen J, Spits H, Roncarolo M, te Velde A, et al. (1991) Interleukin 10 (IL-10) and viral IL-10 strongly reduce antigen-specific human T cell proliferation by diminishing the antigen-presenting capacity of monocytes via downregulation of class II major histocompatibility complex expression. J Exp Med. 174: 915–924.
  22. 22. Soria JC, Moon C, Kemp BL, Liu DD, Feng L, et al. (2003) Lack of interleukin-10 expression could predict poor outcome in patients with stage I non-small cell lung cancer. Clin Cancer Res 9: 1785–1791.
  23. 23. Maehara Y, Kakeji Y, Kabashima A, Emi Y, Watanabe A, et al. (1999) Role of transforming growth factor-beta 1 in invasion and metastasis in gastric carcinoma. J Clin Oncol 17: 607–614.
  24. 24. Hawinkels LJ, Verspaget HW, van Duijn W, van der Zon JM, Zuidwijk K, et al. (2007) Tissue level, activation and cellular localisation of TGF-beta1 and association with survival in gastric cancer patients. Br J Cancer. 97: 398–404.
  25. 25. Wang KS, Hu ZL, Li JH, Xiao DS, Wen JF (2006) Enhancement of metastatic and invasive capacity of gastric cancer cells by transforming growth factor-beta1. Acta Biochim Biophys Sin (Shanghai) 38: 179–186.
  26. 26. Church RD, Yu J, Fleshman JW, Shannon WD, Govindan R, et al. (2004) RNA profiling of cyclooxygenases 1 and 2 in colorectal cancer. Br J Cancer. 91: 1015–1018.
  27. 27. Ristimäki A, Honkanen N, Jänkälä H, Sipponen P, Härkönen M (1997) Expression of cyclooxygenase-2 in human gastric carcinoma. Cancer Res. 57: 1276–1280.
  28. 28. Dubois RN, Abramson SB, Crofford L, Gupta RA, Simon LS, et al. (1998) Cyclooxygenase in biology and disease. FASEB J. 12: 1063–1073.
  29. 29. Fujimura T, Ohta T, Oyama K, Miyashita T, Miwa K (2006) Role of cyclooxygenase-2 in the carcinogenesis of gastrointestinal tract cancers: a review and report of personal experience. World J Gastroenterol. 12: 1336–1345.
  30. 30. Graziano F, Ruzzo A, Santini D, Humar B, Tonini G, et al. (2005) Prognostic role of interleukin-1beta gene and interleukin-1 receptor antagonist gene polymorphisms in patients with advanced gastric cancer. J Clin Oncol. 23: 2339–2345.
  31. 31. Liu J, Song B, Wang JL, Li ZJ, Li WH, et al. (2011) Polymorphisms of interleukin-10 promoter are not associated with prognosis of advanced gastric cancer. World J Gastroenterol. 17: 1362–1367.
  32. 32. Tahara T, Shibata T, Nakamura M, Yamashita H, Yoshioka D, et al. (2011) Effect of IL-1β and TNF-α polymorphisms on the prognosis and survival of gastric cancer patients. Clin Exp Med. 11: 211–217.
  33. 33. Siewert JR, Hölscher AH, Becker K, Gössner W (1987) Cardia cancer: attempt at a therapeutically relevant classification. Chirurg. 58: 25–32.
  34. 34. Lauren P (1965) The Two Histological Main Types Of Gastric Carcinoma: Diffuse And So-Called Intestinal-Type Carcinoma. An Attempt At A Histo-Clinical Classification. Acta Pathol Microbiol Scand 64: 31–49.
  35. 35. Santolaria S, Lanas A, Benito R, Piazuelo E, Sainz R (2001) CagA and VacA cytotoxin antibodies and risk for peptic ulcer disease in patients with Helicobacter pylori infection. Med Clin (Barc) 116: 641–644.
  36. 36. García-González MA, Lanas A, Quintero E, Nicolás D, Parra-Blanco A, et al. (2007) Gastric cancer susceptibility is not linked to pro-and anti-inflammatory cytokine gene polymorphisms in whites: a Nationwide Multicenter Study in Spain. Am J Gastroenterol 102: 1878–1892.
  37. 37. Gonzalez-Conejero R, Rivera J, Corral J, Acuña C, Guerrero JA, et al. (2005) Biological assessment of aspirin efficacy on healthy individuals: heterogeneous response or aspirin failure? Stroke. 36: 276–280.
  38. 38. Cipollone F, Toniato E, Martinotti S, Fazia M, Iezzi A, et al. (2004) A polymorphism in the cyclooxygenase 2 gene as an inherited protective factor against myocardial infarction and stroke. JAMA. 291: 2221–2228.
  39. 39. Cox DG, Pontes C, Guino E, Navarro M, Osorio A, et al. (2004) Polymorphisms in prostaglandin synthase 2/cyclooxygenase 2 (PTGS2/COX2) and risk of colorectal cancer. Br J Cancer. 91: 339–343.
  40. 40. Charlson ME, Pompei P, Ales KL, MacKenzie CR (1987) A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 40: 373–383.
  41. 41. El-Omar EM, Carrington M, Chow WH, McColl KE, Bream JH, et al. (2000) Interleukin-1 polymorphisms associated with increased risk of gastric cancer. Nature 404: 398–402.
  42. 42. Gupta RA, Tejada LV, Tong BJ, Das SK, Morrow JD, et al. (2003) Cyclooxygenase-1 is overexpressed and promotes angiogenic growth factor production in ovarian cancer. Cancer Res. 63: 906–911.
  43. 43. Hwang D, Scollard D, Byrne J, Levine E (1998) Expression of cyclooxygenase-1 and cyclooxygenase-2 in human breast cancer. J Natl Cancer Inst. 90: 455–460.
  44. 44. von Rahden BH, Stein HJ, Pühringer F, Koch I, Langer R, et al. (2005) Coexpression of cyclooxygenases (COX-1, COX-2) and vascular endothelial growth factors (VEGF-A, VEGF-C) in esophageal adenocarcinoma. Cancer Res. 65: 5038–5044.
  45. 45. Chulada PC, Thompson MB, Mahler JF, Doyle CM, Gaul BW, et al. (2000) Genetic disruption of Ptgs-1, as well as Ptgs-2, reduces intestinal tumorigenesis in Min mice. Cancer Res. 60: 4705–4708.
  46. 46. Patrono C, Patrignani P, García Rodríguez LA (2001) Cyclooxygenase-selective inhibition of prostanoid formation: transducing biochemical selectivity into clinical read-outs. J Clin Invest. 108: 7–13.
  47. 47. Takeda H, Sonoshita M, Oshima H, Sugihara K, Chulada PC, et al. (2003) Cooperation of cyclooxygenase 1 and cyclooxygenase 2 in intestinal polyposis. Cancer Res. 63: 4872–4877.
  48. 48. Hou L, Grillo P, Zhu ZZ, Lissowska J, Yeager M, et al. (2007) COX1 and COX2 polymorphisms and gastric cancer risk in a Polish population. Anticancer Res. 27: 4243–4247.
  49. 49. Saxena A, Prasad KN, Ghoshal UC, Bhagat MR, Krishnani N, et al. (2008) Polymorphism of −765G > C COX-2 is a risk factor for gastric adenocarcinoma and peptic ulcer disease in addition to H pylori infection: a study from northern India. World J Gastroenterol. 14: 1498–1503.
  50. 50. Canzian F, Franceschi S, Plummer M, van Doorn LJ, Lu Y, et al. (2008) Genetic polymorphisms in mediators of inflammation and gastric precancerous lesions. Eur J Cancer Prev. 17: 178–183.
  51. 51. Pereira C, Medeiros RM, Dinis-Ribeiro MJ (2009) Cyclooxygenase polymorphisms in gastric and colorectal carcinogenesis: are conclusive results available? Eur J Gastroenterol Hepatol. 21: 76–91.
  52. 52. Zhang X, Zhong R, Zhang Z, Yuan J, Liu L, et al. (2011) Interaction of Cyclooxygenase-2 promoter polymorphisms with Helicobacter pylori infection and risk of gastric cancer. Mol Carcinog. 50: 876–883.
  53. 53. Liu J, Liang Y, Wang ZN, Zhou X, Xing LL (2010) Cyclooxygenase-2 polymorphisms and susceptibility to gastric carcinoma: a meta-analysis. World J Gastroenterol. 16: 5510–5517.
  54. 54. Iglesias D, Nejda N, Azcoita MM, Schwartz SJ, González-Aguilera JJ, et al. (2009) Effect of COX2–765G>C and c.3618A>G polymorphisms on the risk and survival of sporadic colorectal cancer. Cancer Causes Control. 20: 1421–1429.
  55. 55. Coghill AE, Newcomb PA, Poole EM, Hutter CM, Makar KW, et al. (2011) Genetic variation in inflammatory pathways is related to colorectal cancer survival. Clin Cancer Res 17: 7139–7147.
  56. 56. Awad MR, El-Gamel A, Hasleton P, Turner DM, Sinnott PJ, et al. (1998) Genotypic variation in the transforming growth factor-beta1 gene: association with transforming growth factor-beta1 production, fibrotic lung disease, and graft fibrosis after lung transplantation. Transplantation 66: 1014–1020.
  57. 57. Yamada Y, Miyauchi A, Goto J, Takagi Y, Okuizumi H, et al. (1998) Association of a polymorphism of the transforming growth factor-beta1 gene with genetic susceptibility to osteoporosis in postmenopausal Japanese women. J Bone Miner Res 13: 1569–1576.
  58. 58. Castillejo A, Rothman N, Murta-Nascimento C, Malats N, García-Closas M, et al. (2009) TGFB1 and TGFBR1 polymorphic variants in relationship to bladder cancer risk and prognosis. Int J Cancer. 124: 608–613.
  59. 59. Shu XO, Gao YT, Cai Q, Pierce L, Cai H, et al. (2004) Genetic polymorphisms in the TGF-beta 1 gene and breast cancer survival: a report from the Shanghai Breast Cancer Study. Cancer Res. 64: 836–839.
  60. 60. Guan X, Zhao H, Niu J, Tan D, Ajani J, et al. (2009) Polymorphisms of TGFB1 and VEGF genes and survival of patients with gastric cancer. J Exp Clin Cancer Res 28: 94.
  61. 61. Huang RY, Chen GG (2011) Cigarette smoking, cyclooxygenase-2 pathway and cancer. Biochim Biophys Acta. 1815: 158–169.
  62. 62. Shin VY, Cho CH (2005) Nicotine and gastric cancer. Alcohol. 35: 259.
  63. 63. Shin VY, Wu WK, Chu KM, Wong HP, Lam EK, et al. (2005) Nicotine induces cyclooxygenase-2 and vascular endothelial growth factor receptor-2 in association with tumor-associated invasion and angiogenesis in gastric cancer. Mol Cancer Res. 3: 607–615.
  64. 64. Meimarakis G, Winter H, Assmann I, Kopp R, Lehn N, et al. (2006) Helicobacter pylori as a prognostic indicator after curative resection of gastric carcinoma: a prospective study. Lancet Oncol. 7: 211–222.
  65. 65. Hobsley M, Tovey FI, Holton J (2007) Helicobacter pylori and gastric cancer: neither friend nor foe. Gastroenterology. 132: 2076.
  66. 66. Qiu HB, Zhang LY, Keshari RP, Wang GQ, Zhou ZW, et al. (2010) Relationship between H.Pylori infection and clinicopathological features and prognosis of gastric cancer. BMC Cancer. 10: 374.
  67. 67. Gawron AJ, Fought AJ, Lissowska J, Ye W, Zhang X, et al. (2011) Polymorphisms in chemokine and receptor genes and gastric cancer risk and survival in a high risk Polish population. Scand J Gastroenterol. 46: 333–340.