Study On Genetic Susceptibility To Pancreatic Cancer In Chinese Population

Pancreatic cancer (PC) is an aggressive malignancy with extremely low5-year survival rate (<5%) and is the eighth leading cause of the cancer-related mortality worldwide. It is estimated that more than270000new cases diagnosed each year around the world and most of them died within six months. In recent years, the incidence of pancreatic cancer in our country increased rapidly, from relatively rare in the1970s to now the seventh leading cause of the cancer-related mortality. The situation of high fatality of pancreatic cancer has not been improved due to its etiology is not yet clear. At present, surgical resection is the only possible cure method, but because of the lack of effective early diagnostic method, most cases were diagnosed when later period and lost the opportunity of surgery. Malignant tumor is a chronic complicated disease and is the long-term results of interactions of environmental and genetic factors. Pancreatic cancer familial aggregation suggests genetic factors play important roles in the development of PC. Therefore, it is of great significance to carry out studies on genetic susceptibility to PC. The achievements of these studies could be used for specific sensitive biomarkers in early diagnosis, assessment of curative effect and prognosis, and application of new treatment target.Part Ⅰ Study on the associations between genetic variations in SWI/SNF complex and susceptibility to pancreatic cancer Background:Abnormal transcription of genes is the one of the characteristics of cancer. Chromatin is folded in a highly tight status which prevents transcription factors and cofactors access to DNA, so physiological process involving DNA should overcome the high tightness of the chromatin. S WItch/Sucrose Non Fermentable (SWI/SNF) complex is a core factor for chromatin-remodeling that utilize energy of ATP hydrolysis to mobilize nucleosomes, and modulate gene transcription and are widely involved in cell differentiation, proliferation and DNA repair, as a tumor suppressor gene. Recent studies have identified recurrent mutations in major components of SWI/SNF in a variety of human cancers, including PC.Objectives:To explore the associations between genetic variations in SWI/SNF complex and pancreatic cancer risk in Han Chinese; To evaluate gene-gene and gene-environment interactions of genetic variation in SWI/SNF complex in modifying susceptibility of pancreatic cancer.Methods:We carried out a two-stage cases-control study to explore genetic variations in SWI/SNF complex and the risk of PC. For the first stage in Wuhan area, the genotypes were determined using Taqman OpenArray genotyping platforms, and the second stage in the Beijing area, candidate SNPs were genotyped using Taqman genotyping approach. Chi-square test was applied to examine the differences of the distrition of genotypes and demographic data. Armitage’s trend test was used to examine the associations between PC and genotypes. Logistic regression were applied for single SNP analysis and cacluated the OR and95%CI; To control the false positive rate, False discovery rate (FDR) were introduced to control for multiple comparison corrections; Gene-genes and gene-environment interactions are evaluated using logistic regression in both additive and multiplicative model.Results:We conducted a two-stage case-control study to investigate the associations between14common variants in6genes (SMARCA4, SMCRB1, PBRM1, BRD7, ARID1and ARlD2) encoding major components of the SWI/SNF complex and the risk of PC. Three promising variants, rs11644043, rs11085754and rs2073389in the discovery stage comprising310cases and457controls were further genotyped in the validation stage containing429cases and585controls. rs11644043in BRD7and rs11085754in SMARCA4showed consistent significant association with increased risk of PC in both stages, with odds ratios (ORs) and95%confidence interval (CI) of2.04(1.17-3.56) and1.64(1.16-2.33) in stage one, and1.97(1.24-3.14) and1.45(1.04-2.02) in stage two, respectively in a recessive model. Furthermore, the accumulative effects of rs11644043, rs11085754and rs2073389in SMARCB1were observed (P for trend<0.0001). Intriguingly, gene-environmental interactions analysis consistently revealed the potential interactions of rs2073389(Padd-FDR=6.00×10-4, Pmul-FDR=1.50x10-2) and rsl1085754(Padd-FDR=0.03) collaborating with smoking to modify the risk of PC.Conclusion:1. Our comprehensive case-control study identified the main effects of the variants of rs11644043and rs11085754in the SWI/SNF complex on the risk of PC.2. The accumulative effects of the three variants rs11644043, rs11085754and rs2073389in modifying risk of PC were identified.3. The two variants rs11085754, rs2073389interacting with smoking to influence the development of pancreatic cancer.Innovations and limitations:Our study is the first study used two-stage case-control studies to comprehensively investigate the associations between genetic variants in SWI/SNF complex and risk of PC in a Chinese population.The sample size of the second stage is relatively small rendering the lack of enough statistical power. Because we were unable to get the tissue of PC, we could only used the data from the e-QTL database to infer which gene the identified variant influenced, and the level of evidence is relatively low. Part II Study on the associations between genetic variations in9p21.3region and susceptibility to pancreatic cancerBackground:Recently, many GWAS have identified9p21.3as an unexpected hot point, including a variety of metabolic disorders such as coronary artery disease, type2diabetes and several forms of cancer such as breast cancer, glioma, and chronic lymphocytic leukemia. The inactivations of this region were found in approximately30-40%human cancers. Three important human tumor suppressor gene p16INK4a, p14ARE,p15INK4b reside in9p21.3region and they play crucially in regulating the cell cycle inhibitory, aging, apoptosis effect, and widely participate in the development of various kinds of cancers. Another important gene in this area is ANRIL, a long non-coding RN A and nowadays it becomes very attractive to scientific community because it has been found to play a vital role in regulating the expression of p161NK4a and p15INK4b through a cis-acting mechanism. Multiple tumor suppressor genes of this area play important roles in the development of PC, at present, there is no study to explore the associations between genetic variations in this area and PC risk in Chinese population.Objectives:To explore the associations between genetic variations in9p21.3region and PC risk in Han Chinese; To examine the biological function of genetic variations within9p21.3associated with susceptibility to PC.Methods:We carried out a two-stage cases-control study to explore genetic variations in9p21.3and the risk of PC. For the first stage, we selected the tagSNPs in9p21.3(19.9-29MB) and used the genotype data from our completed exons chip project of PC. Chi-square test and logistic regression were applied for single SNP analysis and false discovery rate (FDR) were introduced to control for multiple comparison corrections. We used VISTA Enhancer Browser database, chip-seq data from Encode and SNPexp-A QTL database for bioinformatics analysis to find potential functional variants which identified in the first stage or LD with those positive variants. We further explored potential biological functions of the SNPs using Dual luciferase reporter assay and the electrophoretic mobility shift assay (EMSA).The potentially functional SNPs were further genotyped using Taqman genotyping method in stage two in Wuhan area;Results:1. For the first stage,942pancreatic cancer and1504controls were finally included, and we analyzed39common variants in9p21.3region. We found rs6475609in ANRIL was associated with a decreased risk of PC (OR=0.81,95%CI=0.72-0.92). Using bioinformatics analysis and association study we found two potential functional variants rs1537373(OR=0.84,95%CI=0.74-0.95) and rs1333042(OR=0.84,95%CI=0.74-0.95) which are LD with rs6475609were also associated with a decreased risk of PC. We successfully validated the variant rs1537373(OR=0.84,95%CI=0.73-0.98) with potential function in stage two in624pancreatic cancer patients and1244controls in Wuhan area.2. The results of e-QTL indicated the genotypes of rsl537373were associated with expression of CDKN2B gene, individuals carrying TG/TT genotypes exhibited higher level of CDKN2B expression than individual carrying GG genotype (P dominant=6.00×10-4).3. Dual luciferase reporter assay indicated that fragment with rs1537373-G showed higher activation of enhancer than fragment with rs1537373-T in two pancreatic cancer cell lines Panc-1(P<0.01) and SW1990(P<0.01).4. Electrophoretic mobility shift assay indicated fragment with rs1537373-G showed stronger ability of combination with certain transcription factors than fragment with rs1537373-T.Conclusion:1. Rs1537373in ANRIL gene in9p21.3region contributed significantly to the risk of PC.2. Rs1537373might modify the risk of PC through different ability of combination with certain transcription factors and thus influence the enhancer activation of the fragment.Innovations and limitations:Our study is the first study used two-stage case-control studies intergrating biological function study to comprehensively investigate the associations between genetic variants in9p21.3region and risk of PC in a Chinese population. Our study devoted much efforts in explore the effects of gene-gene and gene-environmental interactions which have been usually ingnored before.The sample size of current study is relatively small, especially when evaluated the gene-gene and gene-environmental interactions. The variants identified in current study have not been verified by biological functional study, so the underlying mechanism of these variants is unclear.