Study On The Association Between Family History Of Tumors And Gastric Cancer Risk

Background and ObjectivesGastric cancer(GC)is one of the most common malignant tumors in the world.From the detection data of GC in recent years,the incidence of GC has been decreasing,particularly in developed countries.But it also ranks as the 5th most common cancer worldwide and the 3rd leading cause of cancer mortality.Owing to the aging of the populations,the absolute number of GC cases is increasing and the disease burden of GC is heavy.In a global context,East Asia(especially Korea,Japan and China),Central and Eastern Europe and South America are high incidental areas of GC;while,North America and Africa are areas with low incidence of GC.China is in a region with a high incidence of GC.Focus on China,there is still a significant regional difference of GC incidence.Shanghai,Gansu,Jiangsu,Qinghai are provinces with high GC incidence,and Taixing County of Jiangsu province is also included.The latest report of malignant tumors in Taixing,Jiangsu(2013)indicated that the overall incidence of GC was 38.5/100,000 and the mortality rate was 32.8/100,000.And the overall incidence of GC in Taixing is higher than the national average level(31.28/100,000).Moreover,GC has been ranked 3rd in terms of tumor incidence and mortality,after esophageal cancer and liver cancer in this area.According to the traditional classification standard(Lauren pathological classification standard),GC can be divided into two major categories,intestinal type and diffuse type.The Cancer Genome Atlas(TCGA)molecular classification divided GC into four subtypes:Epstein-Barr virus related tumors,microsatellite unstable tumors,genetically stable tumors and chromosomal instability tumors.The onset and development of GC is the result of long-term effects of endogenous factors(genetic factors)and exogenous factors(chemical and biological factors).Helicobacter pylori(HP)infection,high-salt or high-oil diet,Epstein-Barr virus(EBV)infection and E-cadherin(CDH1)mutations are considered as important risk factors for GC.Although there is a certain understanding of the risk factors of GC,the etiology of GC is still not clear.Coupled with the heavy disease burden of GC,it is necessary to further explore the etiology of GC.All diseases have their own genetic basis during their onsets and developments,and GC is no exception.Genetic factors are usually measured and studied through family history of tumors aiming at enhance the practical significance of the conclusion.With regard to the studies on the relationship between family history of tumors and GC,we can date back to 1979.XuXinggao used the method of drawing the pedigree map of GC to demonstrate the familial aggregation of GC.After that,there were several studies to explore the association between family history of tumors and GC risk conducted in Dalian,Henan Lin County,Gansu Wuwei City,Shanxi Xianyang City and other high-risk areas of GC in our country.Their conclusions all suggested that the family history of tumors plays an important role on the onset of GC.But these previous studies mostly used cross-sectional study or case-control study design.Small sample size,inconsistency subgroup analyses results and not distinction between GC and gastroesophageal junction tumors were general limitations in these studies.In addition,data from kin cohorts are still scarce.Therefore,it is necessary to discuss the relationship between family history of tumors and the incidence of GC in depth.From 2010 to 2013,our research group had carried out a population-based case-control study on the etiology of upper digestive tract tumors in Taixing.During the survey,we found that GC exhibits family aggregation in this area,providing basis of research.So we used part of the information collected from this population-based case-control study in Taixing to explore the relationship between the family history of tumors and GC risk by two different designs(case-control study and kin-cohort study)in order to provide some scientific support for the establishment of early screening of GC in the local residents and provide reference for future research on genetics of GC.Subjects and MethodsBased on the strict population-based case-control study,we collected new GC cases from four hospitals which could do gastroscopy during 2010.9 to 2013.10.And those missed cases were supplemented from Cancer Registry inTaixing Center for Disease Control and Prevention at the end of each year.Controls were randomized chosen by category matching conditions from Taixing County Population Registration System.All subjects were interviewed face-to-face using a standardized questionnaire.After cleaning up this part of family history information,we used case-control design and kin-cohort design to analyze the relationship between the family history of tumors and GC risk,respectively.In the case-control study design,the association between the family history of tumors and GC risk was explored by unconditional logistic regression analyses,using odds ratio(OR)and 95%confidence interval(95%CI).Subgroup analyses were made according to the different tumor types and the number of first-degree relatives(FDRs)suffering from tumors.In the kin-cohort study design,FDRs of our probands(case group and control group)formed a kin cohort.Then we followed up this kinship cohort until the study outcomes(incidence,death or lost to follow-up)appeared.Meanwhile we recorded their incidence age of tumors or the age of death.According to whether the proband suffered from GC or not,this kinship cohort was divided into exposure group and non-exposure group.First,Kaplan-Meier survival curve and cumulative risk curve of Nelson-Aalen were drawn respectively.Then Cox proportional hazards models were established in order to compare the cumulative risk of upper gastrointestinal cancer and GC by the age of 85.All analyses were finished by Stata14.0.Results1.Case-control study.A total of 481 incident GC cases and 1972 matched controls are identified in this study.Family members,distributions of compatriots and offspring were balanced and comparable between case group and control group.The positive rate of GC family history in Taixing was 11.0%in the control group and 20.0%in the case group.Results are as follows:①Family history of any type tumor in any FDRs was associated 1.64times the GC risk compared to subjects without family history of GC(OR=1.64,95%CI:1.33-2.02).Any type of cancer in one or both parents was associated with an OR of 1.44 compared to those with no parental history of any cancer(OR=1.44,95%CI:1.15-1.81).A positive tumor family history in siblings increased GC risk(OR=1.60,95%CI:1.27-2.03).②Family history of any digestive system cancers in any FDRs increased the risk of GC by 69%(OR=1.69,95%CI:1.37-2.08).A positive family history of esophageal cancer was associated a higher risk of GC by 41%(OR=1.41,95%CI:1.08-1.84).But there is no relationship between family history of colorectal cancer and GC risk(OR=1.75,95%CI:0.96-3.17).③After adjustment for age and sex,family history of GC in any FDRs increased GC risk compared to those without family history of GC in any FDRs(OR=2.10,95%CI:1.61-2.73);these associations remained significant after adjustment for multiple baseline covariates(OR=2.11,95%CI:1.60-2.78).Furthermore,with the more FDRs suffered from tumors the risk of GC increased(P<0.001).GC in one or both parents was associated with 2.19-times compared to those without parental history of GC(OR=2.19,95%CI:1.58-3.02).It was also significant after adjustment for multiple baseline covariates(OR=2.20,95%CI:1.58-3.08).Considered separately,GC risk of paternal history of GC was 2.01-fold compared to those without paternal history of GC;whereas maternal history of GC was 2.75-fold compared to those with no maternal history of GC.GC risk with sibling history increased(OR=1.89,95%CI:1.26-2.84).2.Kin-cohort study.There were 15744 FDRs of 1972 controls in non-exposed group and 3937FDRs of 481GC cases in exposed group.Up to the age of 80,the cumulative risk of GC was 9.48%for FDRs and 6.64%for FDRs of controls.Up to the age of 85,after adjustment for gender,the cumulative risk of getting upper gastrointestinal tumors(GC and esophageal cancer)in the FDRs of cases was 1.44-fold compared to those FDRs of controls(HR=1.44,95%CI:1.25-1.65).Results of subgroup analyses were that the cumulative risk of upper gastrointestinal tumors in male FDRs of case group was 38%higher than in male FDRs of control group(HR=1.38,95%CI:1.17-1.63).As to females,we got a HR of 1.58(HR=1.58,95%CI:1.24-2.02).In addition,the cumulative risk of upper gastrointestinal tumors in siblings of case group was 67%higher than in siblings of control group(HR=1.67,95%CI:1.39-2.00).Considered GC risk only,we found that up to the age of 85,the cumulative risk of getting GC in the FDRs of cases was 1.81-fold compared to those FDRs of controls(HR=1.81,95%CI:1.47-2.22).After adjustment for gender,the result was HR(95%CI)=1.79(1.46-2.20).Subgroup analyses indicated that the cumulative risk of GC in male FDRs of case group was 64%higher than in male FDRs of control group(HR=1.64,95%CI:1.29-2.08).While we got a HR of 2.37 in female FDRs(HR=2.37,95%CI:1.59-3.55).In addition,the cumulative risk of GC in siblings of case group was 71%higher than in siblings of control group(HR=1.71,95%CI:1.29-2.27).Conclusion1.Positive family history of tumors is a risk factor for GC.With the more FDRs in family suffered from tumors the risk of GC increased.2.Different locations of digestive tract cancer(esophageal cancer,GC,colorectal cancer)suffered from in FDRs have different GC risk.3.The cumulative risk of getting GC in the FDRs of GC patients is higher than that in the FDRs of healthy population.