Research On Etiology And Population Attributable Of Environment Risk Factors Of Gastric And Esophageal Cancers

Objectives:1.To study the association between of serum Hp infection and pepsinogens with risk ofgastric non-cardia adenocarcinoma (GNCA), gastric cardia adenocarcinoma (GCA),and esophageal squamous cell carcinoma (ESCC) in Linxian,China.2.To estimate the contribution of tobacco smoking, alcohol drinking, infection, lowvegetable and rfuit intake to the mortality and incidence of gastric and esophagealcancers in China.Material and Methods:1.Research on etiology: In a cohort of22033Linxian Nutrition Intervention Trial(NIT) surviving participants in NIT follow ups rfom1999to2007,we performed aprospective nested case-control study. Cases were selected rfom the random samplesof the diagnosed GNCA, GCA and ESCC cases by2007. Controls were collectedrfom the random samples of noncancer subjects. We assessed serum Hp IgGantibodies, pepsinogen I and II by using ELISA. Logistic regression models withadjustment for sex, age, smoking, dirnking, body mass index and commune was usedto estimate the odds ratios (OR) and95%confidence intervals (95%CI).2.Research on PAF: We used the findings of systematic review of prevalence ofknown risk factors,including tobacco smoking, alcohol drinking, infection, lowvegetable intake and low rfuit intake and their relative risks,and calculated theproportion of gastric and esophageal cancers attributable to known modifiable riskfactors (population attributable fraction(PAF)). PAF is defined as the proportionalreduction in average disease risk over a specified time that would be achieved byeliminating certain causal factors. Exposure data was taken from meta-analysesand large-scale national surveys of representative samples of the Chinesepopulation. Data on relative risks were from meta-analyses and large-scaleprospective studies. Cancer mortality and incidence data came rfom the3rdnational death cause survey and population-based cancer registries in China. Results:1.Research on etiology:1)Among control group,68.3%of the participants were infected with Hp; Hpinfection increased the level of serum PGI and PGII, but decreased PGR.2)Hp infection was a risk factor for GNCA, adjusted ORs (95%CI) were1.58(1.01-2.48), low PGI (<50|ig/L)and PGR (<4) were associated with GNCA,adjusted ORs were2.36(1.14-4.88) and1.98(1.22-3.20) respectively. Thecombined bio marker of low PGI and PGR had a stronger association with GNCA(ORs=2.39).3)Hp infection and low PGR (<50|ig/L) were risk factors for GCA, adjusted ORs(95%CI) were1.41(1.04-1.91) and1.45(1.06-1.99) respectively, low PGI wasnot a risk factor of GCA.4)Both Hp infection and low serum PG were not associated with ESCC.2.Research on PAF:1)We estimated that223768GNCA deaths (men:148169; women:75599) and300608cases (men:203719, women:96889) were attributable to tobaccosmoking, infection,low vegetable intake and low fruit intake in China in2005.About30.9%of GNCA deaths among men and3.8%among women wereattributable to tobacco smoking. About63.1%of GNCA deaths were caused byinfection. Low vegetable intake was responsible for16.9%GNCA deaths in menand16.1%in women. The rfaction of GNCA deaths attributable to low fruitintake was38.6%in men and39.8%in women. Overall,85%of GNCA (87%inmen and82%in women) were attributable to these four modifiable risk factors.2)We estimated that38821GCA deaths (men:30461; women:8360) and52595cases (men:41880, women:10715) were attributable to tobacco smoking,infection, low vegetable intake and low rfuit intake in China in2005. About30.9%of GCA deaths among men and3.8%among women were attributable totobacco smoking. About19.2%of GCA deaths were caused by infection. Lowvegetable intake was responsible for16.9%GCA deaths in men and16.1%inwomen. The rfaction of GCA deaths attributable to low rfuit intake was38.6%inmen and39.8%in women. Overall,69%of GCA (72%in men and61%inwomen) were attributable to these four modifiable risk factors. 3) We estimated that85421EC deaths (men:66066; women:19355) and106167cases (men:81958, women:24209) were attributable to tobacco smoking, alcohol drinking, low vegetable intake and low fruit intake in China in2005. About17.9%of EC deaths among men and1.9%among women were attributable to tobacco smoking. About15.2%of EC deaths in men and1.3%in women were caused by alcohol drinking. Low vegetable intake was responsible for4.3%EC deaths in men and4.1%in women. The fraction of EC deaths attributable to low fruit intake was27.1%in men and28%in women. Overall,46%of EC (50%in men and33%in women) were attributable to these four modifiable risk factors.Conclusions:1. Control group had a high prevalence of Hp infection, Hp infection increased the level of serum PGI and PGII, but decreased PGR.2. Hp infection, low PGI and low PGR were risk factors for GNCA, The combined bio marker of low PGI and PGR had a stronger association with GNCA than PGI or PGR alone. The bio markers of serum PG and Hp or PG alone for GNCA screening were not recommended.3. Hp infection and low PGR were risk factors for GCA, low PGI was not a risk factor of GCA.4. Both Hp infection and low serum PG were not associated with ESCC.5. Tobacco smoking, infection, low vegetable intake and low fruit intake contributed to85%GNCA and69%GCA mortality and incidence in China in2005. Tobacco smoking, alcohol drinking, low vegetable intake and low fruit intake were responsible for46%(50%in men and33%in women) of esophageal cancer mortality and incidence. These findings provide useful data for developing guidelines for gastric cancer and esophageal cancer prevention and control in China.