| About one million new cases of gastric cancer were estimated to have occured in2008（7.8%of the total new cancer cases）, making it the fourth most commonmalignancy in the world, behind cancers of the lung, breast and colo-rectum. Morethan70%of gastric cancer cases occur in developing countries, and half the worldtotal occurs mainly in China. Gastric cancer is also the second leading cause of cancerdeath worldwide. Despite the incidence and mortality of gastric cancer keep decliningover past decades worldwide, they still remain high in China. So disease preventionand control in gastric cancer is still a key public health issue in China.Gastric cancer is anatomically divided into cardia gastric and non-cadria gastriccarcinoma. Gastric tumors located within20mm distal to the gastroesophagealjunction are defined as cardia gastric cancer, and those which origin from otheranatomical locations of stomach are non-cardia gastric cancer. It has been shown thatgastroesophageal reflux and obesity contribute to the risk of cardia gastric cancer. Incontrast, Helicobacter pylori infection is the major etiological factor for non-cardiagastric cancer but not for cardia, indicating the potential difference in mechanisms ofthese two types of gastric cancer. Gastric cancer derives from a complexmultiple-stage carcinogenesis, including chronic atrophic gastritis, intestinalmetaplasia, dysplasia and final gastric cancer. The exact etiology of gastric cancer is still unclear, both environmental and genetic factors contribute to it. Although theprevalence of H. pylori infection ranges from40%to80%in humans, only a smallproportion of infected individuals （1%） develop gastric cancer, suggesting individualshave different cancer susceptibility even under the same risk factors exposure. Thissusceptibility is to be determined by genetics, which is mainly focused on SingleNucleotide Polymorphisms （SNPs）. So it’s valuable in detecting genetic susceptibilitymarkers to further explore etiology and mechanism of gastric cancer and the potentialapply in disease screening, diagnosis and therapy.In2008, the first gastric cancer GWAS identified SNP rs2976392onchromosome8q24loci as a susceptibility SNP for diffuse gastric cancer (OR=1.71,P=1.5×10-16). Meanwhile, different distributions (P=6.0×10-4) of SNP rs2976392indiffuse and intestinal gastric cancer indicating genetic heterogeneity exists betweensubtypes of gastric cancer. In2010, Abnet et al. reported that SNP rs2274223in10q23loci was associated with cardia gastric cancer (screening OR=1.59,P=5.88×10-14, validation OR=1.49, P=0.016, combined OR=1.57, P=4.19×10-15) butnot with non-caidia gastric cancer （screening OR=1.02, P=0.78, validation OR=1.10,P=0.34, combined OR=1.05, P=0.44）. At the same time, Wang et al. confirmed thesame loci for cardia gastric cancer (OR=1.55, P=1.74×10-39) in another independentGWAS. According to these two publications,10q23was a specific susceptibility locifor cardia gastric cancer but not for non-cardia. This finding strongly suggested thegenetic heterogeneity in cardia and non-cardia gastric cancer. Now10q23loci isnewly proved cardia gastric cancer specific susceptibility loci, while the counterpartsfor non-cardia are totally unknown. So we conduct a well-designed non-cardia gastriccancer to identify some specific susceptibility loci for non-cardia gastric cancer.The project was a3-stage study, with GWAS screening （1,006non-cardiagastric cancer cases and2,273health controls） on Affymetrix GeneChip Human Mapping6.0platform and validation I （1,894cases and2,090controls） and validationII （1,394cases and1,519controls） on TaqMan Genotyping Assays. Screening stageincluded Nanjing and Beijing two independent studies （565cases and1,162controlsfrom Nanjing,468cases and1,123controls from Beijing） with both independent andcombined screening. Association analyses were evaluated by logistic regression ingenetic additive model. Those SNPs with strong association signals were replicated invalidation I study, and consistent ones were further tested in validation II study. Wealso evaluated the relationship between finally identified SNPs and cardia gastriccancer in905cardia gastric cancer case and3,609shared controls. In addition, weconducted imputation analysis to assess previously GWAS reported gastric cancersusceptibility loci for non-cardia gastric cancer in our study.We found17loci with strong association signals with non-cardia gastric cancerin GWAS screening analysis, whose P value were from6.52×10-6to3.47×10-11, andodds ratio （OR） ranged from0.47to1.44. Among these17loci in validation I study,3q13.31loci （rs9841504C>G in ZBTB20region: OR=0.81, P=1.94×10-3） and5p13.1（rs13361707T>C in PTGER4-PRKAA1region: OR=1.38, P=2.82×10-12） were wellreplicated. And these two SNPs were further consistent confirmed in Validation IIstudy （rs9841504C>G: OR=0.78, P=3.4×10-3; rs13361707T>C: OR=1.46,P=6.9×10-12）. Combined screening and validation analysis revealed that SNPrs9841504G allele could decrease24%per-allele non-cardia risk （OR=0.76,P=1.7×10-9） while SNP rs13361707C allele could increase41%per-allelenon-cardia risk（OR=1.41, P=7.6×10-29）. In stratified analysis by age, sex, smokingand drinking status, rs9841504and rs13361707were significant associated withnon-cardia gastric cancer risk in each sub-group, none heterogeneity was detected ineach stratified variable. Cardia gastric cancer association study suggested rs9841504（OR=1.00, P=0.968） and rs13361707（OR=1.12, P=0.054） were not associated with cardia gastric risk. Imputation analysis confirmed previously gastric cancer GWASreported1q22（rs4072037: OR=0.73, P=1.7×104）,8q24（rs2294008: OR=1.36,P=2.1×107; rs2976392: OR=1.35, P=3.7×107） and20p13（rs13042395: OR=0.80,P=6.9×104） were associated non-cardia gastric cancer susceptibility, while10q23（rs2274223: OR=1.11, P=0.136） was not non-cardia gastric cancer susceptibility lociin our population.In this study, We identified two new specific susceptibility loci fornon-cardia gastric cancer at5p13.1and3q13.31, and replicated the associationbetween genetic variants at1q22,8q24,20p13and non-cardia gastric risk in ourpopulation. It’s valuable that our findings in detecting more susceptibility factors andillustrating potential mechanisms on gastric cancer. Meanwhile, our newly identifiedsusceptibility SNPs as well as other GWAS reported markers might be applied innon-cardia gastric cancer high risk population screening and promising moleculartherapy targets in gastric cancer treatment. In summary, our findings are veryimportant in understanding of the etiology and mechanisms of gastric cancerdevelopment, which in turn is important for early detection, diagnosis and targetedtreatment of this malignancy. |
PDF Full Text Request