福建省汉族人群肺癌相关microRNA基因SNP与吸烟交互作用研究

Interaction research on smoking and microRNA genes SNP related to lung cancer in Fujian Han population

目的探讨福建省汉族人群肺癌相关microRNA（miRNA）基因SNP位点rs11614913、rs2910164、rs12894467、rs7372209及rs895819与吸烟的交互作用关系。方法采用病例一对照研究设计，收集2006年1月至2012年1月经病理确诊的汉族新发原发『生肺癌患者1053例；按照性别、年龄进行频数成组匹配，同期选取探视病例组患者的汉族亲友及在福建省福州市苍霞社区卫生服务中心进行健康体检的汉族人群作为对照组，共1058名。调查研究对象性别、身高、体重、文化程度、婚姻状况、肿瘤家族史、肺部疾病史、吸烟、饮茶、饮酒等情况。经知情同意后，均采集空腹静脉血5ml，应用基质辅助激光解析电离时间飞行质谱技术（MALDI-TOF-MS）进行rs11614913、rs2910164、rs12894467、rs7372209及rs895819位点多态基因分型。以是否发生原发性肺癌为因变量，以SNP位点为白变量构建多因素非条件logistic回归模型。利用叉生分析探讨SNP位点与吸烟之间可能存在的联合作用；利用超额相对危险度（RERI）分析吸烟与SNP位点显、隐性模型的相加交互作用。结果病例组吸烟量P50（P25～B75）为30．00（0．00～56．00）包年，高于对照组[0．00（0．00～20．48）包年1（Z=14．57，P〈0．001）；病例组不吸烟者的被动吸烟指数P50（P25～P75）为11．40（0．00～25．00），高于对照组[0．00（0．00～13．11）1（Z=10．71，P〈0．001）。rs11614913、rs2910164、rs12894467、rs7372209及rs895819位点检出率在病例组中分别为95．82％（1009／1053）、97．72％（1029／1053）、97．82％（1030／1053）、97．15％（1023／1053）和96．01％（1011／1053）；对照组中分别为98．11％（1038／1058）、98．96％（1047／1058）、98．30％（1040／1058）、98．68％（1044／1058）和98．02％（1037／1058）。rs11614913位点的显性遗传模型中，TT基因型与吸烟联合可增加原发性肺癌的发病风险（OR=4．04，95％CI：2．67～6．12）；隐性遗传模型中，CC基因型与吸烟联合可增加原发性肺癌的发病风险（OR=4．76，95％CI：3．16～7．17）。rs12894467位点的显性遗传模型中，TT基因型与吸烟联合可增加原发性肺癌的发病风险（OR=2．98，95％CI：2．28～3．90）；隐性遗传模型中，CC基因型与吸烟联合可增加原发性肺癌的发病风险（OR=1．94，95％CI：1．10～3．43）。rs2910164位点的显性遗传模型中，CC基因型与吸烟联合可增加原发性肺癌的发病风险（OR=2．18，95％CI：1．60～2．98）；隐性遗传模型中，GG基因型与吸烟联合可增加原发性肺癌的发病风险（0R=3．29，95％CI：2．16～5．03）。吸烟与rs12894467位点显、隐性基因模型相乘交互项的联合作用均存在统计学意义（x^2=8．58，P=0．003；x^2=4．76，P=0．040）。结论rs11614913、rs12894467及rs2910164位点多态性与福建省汉族原发性肺癌发生存在潜在关联。

Objective To investigate the interaction on smoking and the lung cancer related genes miR-196a2 rs11614913, miR-146a rs2910164, miR-300 rs12894467, miR-26a-1 rs7372209, miR-27a rs895819 in Fujian Han population. Methods From January 2006 to January 2012, by using a hospital-based case-control study, 1 053 cases were pathologically diagnosed as primary lung cancer from the Department of Thoracic Surgery and 1 058 controls were randomly selected from the visiting relatives of patients and visiting people of Cangxia community health service of Fuzhou city according to match with age and genders. They were recruited for questionnaires survey and genotyping detection. Research objects of genders, height, weight, cultural degree, marital status, family history of cancer, lung disease history, smoking, drinking tea, drinking, and so on. After informed consent, we collected 5 ml fasting venous blood from every object, used MALDI-TOF-MS to analysis genotyping of polymorphic loci. Logistic regression model was constructed by using SNP as independent variable, and the multiple factors were constructed with different loci. The possible association between SNP and cigarette smoking was analyzed by using the crossover analysis.＇ The relative excess risk of interaction （RER1） were used to analyze on smoking and SNP loci additive interaction of dominant and recessive genetic models. Results Smokers in case group who smoked P50（P25-P75）30.00 （0.00-56.00） packages in a year were higher than control group （0.00（0.00 - 20.48） pack years） （Z=14.57, P〈0.001）. Passive smoking index for non-smokers was 11.40（0.00-25.00）, higher than the controls （0.00（13.00- 13.11）） （Z=10.71, P〈0.001）. Site detection rate of rs11614913, rs2910164, rs12894467, rs7372209 and rs895819 in cases was 95.82%（1 009/1 053）, 97.72%（1 029/1 053）, 97.82% （1 030/1 053）, 97.15% （1 023/1 053） and 96.01% （1 011/1 053） respectively. The controls were 98.11% （1 038/1 058）, 98.96% （1 047/1 058）, 98.30% （1 040/1 058）, 98.68% （1 044/1 058） and 98.02% （1 037/1 058） respectively, rs11614913 dominant genetic model, TT genotype and smoking could increase the risk of primary lung cancer （OR=4.04, 95%CI：2.67 - 6.12）. Recessive genetic model, CC genotype and smoking increased the incidence of primary lung cancer risk （OR=4.76, 95%CI： 3.16 - 7.17）. rs12894467 dominant genetic model, TT genotype and smoking could increase the risk （OR=2.98, 95%CI： 2.28 - 3.90） in primary lung cancer. In recessive genetic model, CC genotype and smoking increased the incidence of primary lung cancer risk （0R=1.94, 95% CI： 1.10-3.43）. Dominant genetic model of rs2910164, CC genotype and smoking could increase the risk （OR=2.18, 95%CI： 1.60 - 2.98） in primary lung cancer. Recessive genetic model, GG genotype and smoking increased the incidence of primary lung cancer risk （OR=3.29, 95%CI： 2.16 - 5.03）. Especially rs12894467 dominant and recessive gene model and genders, smoking and there were combined effects（x^2=8.58, P=0.003; x^2=4.76, P=0.040）. Conclusion Rs11614913, rs12894467 and rs2910164 polymorphism were potentially associated with primary lung cancer in Fujian Han population.