【作者】 张爽爽；

【导师】 陈坤；

【作者基本信息】 浙江大学 ， 劳动卫生与环境卫生学， 2010， 博士

【摘要】 研究背景结直肠癌是人类常见的消化道肿瘤之一,在西方国家发病率较高。虽然我国是结直肠癌的传统低发国家,但近几年来随着人民生活水平的提高,饮食习惯的西化,我国结直肠癌的发病率和死亡率呈逐年上升趋势。2005年我国的结直肠癌年龄标化发病率男性和女性分别达到15.0／10万和9.7／10万,居常见恶性肿瘤发病率第5位。结直肠癌的发生发展是一个多因素、多步骤、多阶段的过程,是外在的环境因素和机体内在的遗传因素相互作用的结果。而环境因素至少可以解释70%的散发性结直肠癌的发生。其中饮食(高脂肪、高热量、低纤维素饮食)、吸烟和饮酒等生活方式、超重与肥胖、肠道疾病史等可能起着重要的作用。环境中持久性有机污染物对环境和人体健康影响的研究已成为当今环境与卫生科学研究的前沿,受到国际社会的高度重视。2001年联合国环境规划署通过的《关于持久性有机污染物(POPs)的斯德哥尔摩公约》明确指出禁止使用或严格限制使用滴滴涕、六六六、多氯联苯等毒性极强且难降解的化学物品。由于这类物质化学性质稳定,残留时间长,能造成环境和动植物体内的大量积累,且可通过生物富积和食物链作用,蓄积于人体内。动物实验证明有机氯对鼠类致癌率很高,流行病学上也已经将有机氯与多种癌症联系起来。有报道表明,有农药接触史、血清中农药水平、职业接触有机氯农药、饮用氯化消毒的水源等都可能与结直肠癌死亡(发病)有关联。因此,本研究以浙江省嘉善县为研究现场,采用匹配的病例对照研究设计,同时以血浆中的有机氯化合物(包括α-HCH、β-HCH、γ-HCH、δ-HCH；P-p’DDT、P-p’DDE；PCB28、52、101、118、138、153和180)为测量指标,探讨血浆中有机氯化合物与结直肠癌易感性的关系,以及有机氯化合物分别和环境因素、ⅠⅡ相代谢酶基因多态的联合作用与结直肠癌易感性的关系。材料与方法采用病例对照研究设计,以全国结直肠癌调整死亡率最高县---浙江省嘉善县为研究现场,以该县1990年4月建立的包括10个乡镇的当地居民(64 693人)随访队列为研究人群。以队列内1990年5月至2005年5月发生的,且曾经参加过1989-1990年开展的结直肠癌普查的原发性结直肠癌存活病例,排除直肠类癌、继发性结直肠癌等,同时排除拒访、失访,最终纳入结直肠癌病例206例(结肠癌93例,直肠癌113例)。在同一队列人群中,按同居住地、同性别、年龄±5岁1：2个体匹配抽取对照,抽取了412例健康人群组成对照组。因血浆样品丢失或者实验室检测失败,最终纳入分析的有效病例数202例,对照404例。通过问卷调查收集个体特征和生活方式等常见环境暴露因素的信息,经调查对象知情同意,采集静脉血5ml,分装储存备用。现场调查的质量控制通过调查前统一培训调查员,调查中统一携带标准的调查员手册,调查后统一对问卷进行复核和以5%的样本量进行电话回访实现。采用液相微萃取技术分离血浆中的有机氯化合物,同时运用气相色谱—电子捕获检测技术(GC-ECD)分别测定各有机氯化合物的含量。调查表经统一编码,EPIDATA软件双遍输入,校对无误后进行统计分析。分析过程主要包括：婚姻状况、职业、文化程度、BMI等人口学因素的均衡性采用Pearson x2检验；研究因素的比值比(OR)值和95%可信区间(CI)估计采用Cox回归方法；对非正态的有机氯化合物含量,按对照组33.3%和66.6%分布,分为三个等级,以最低暴露量组为参照组；常见生活方式因子与有机氯化合物的交互作用、代谢酶基因多态性与有机氯化合物的交互作用采用又生分析,交互P值采用似然比检验计算。结果1常见生活方式因子与结直肠癌人口学特征在病例组和对照组中的分布差异均没有统计学意义。吸烟史、饮酒史和饮茶史与结直肠癌的发病风险没有统计学关联。但是,相对于无吸烟史者,吸烟年限>35年者结直肠癌发病风险增加并有统计学意义(OR：2.20,95%CI：1.13-4.27),但是趋势检验P值为0.13,没有统计学意义。2有机氯化合物与结直肠癌血浆中有机氯化合物(包括α-HCH、β-HCH、γ-HCH、δ-HCH；P-p’DDT、P-p’DDE；PCB28、52、101、118、138、153和180)在病例组和对照组中含量的分布差异没有统计学意义,但是经分层发现,经年龄、性别、婚姻状况、职业、BMI等因素校正后,高暴露组(>4.03 ng/g)的γ-HCH、高暴露组(>1157.61ng/g)的P-p’DDE、高暴露组(>36.79ng/g)的PCB118、高暴露组(>8.09ng/g)的PCB180以及高暴露组(>341.53ng/g)的Total-PCBs使结直肠癌的发病风险均增加,且都有统计学意义,OR值分别为1.36(95%CI：1.04-1.77)、1.31(95%CI：1.01-1.71)、1.36(95%CI：1.05-1.67)、1.45(95%CI：1.06-1.99)和1.35(95%CI：1.00-1.82)。3常见生活方式因子、有机氯化合物的联合作用与结直肠癌高暴露组的PCB138和有吸烟史增加结直肠癌的发病风险,OR值为1.96(95%CI：1.19-3.23)。吸烟年限>35年,分别和高暴露组的α-HCH、P-p’DDT、P-p’DDE、PCB52、138、180以及总PCBs的联合作用都能增加结直肠癌的发病风险,OR值分别为2.00(95%CI：1.09-3.65)、1.82(95%CI：1.03-3.22)、1.95(95%CI：1.11-3.43)、1.92(95%CI：1.08-3.44)、1.93(95%CI：1.02-3.66)、2.61(95%CI：1.33-5.11)和2.44(95%CI：1.35-4.41)。另外,PCB180和吸烟年限存在有统计学意义的交互作用(P=0.04)。吸烟量>20支／天,分别和高暴露组的总HCH、P-p’DDE、PCB118、180以及总PCBs的联合作用能增加结直肠癌的发病风险,OR值分别为1.82(95%CI：1.00-3.32)、2.03(95%CI：1.14-3.60)、2.08(95%CI：1.13-3.83)、2.72(95%CI：1.41-5.23)和1.92(95%CI：1.08-3.39)。饮酒史与各有机氯化合物(包括六六六、滴滴涕和多氯联苯)联合作用对结直肠癌发病风险的影响没有统计学意义。但是,高暴露组的PCB52和饮酒年限>31年、高暴露组的PCB180和饮酒年限>31年均增加结直肠癌的发病风险,OR值分别为1.94(95%CI：1.00-3.78)和1.77(95%CI：1.05-3.28)。低暴露组的P-p’DDT和有饮茶史的联合作用能降低结直肠癌的发病风险,且有统计学意义,OR值为0.59(95%CI：0.37-0.94)；高暴露组的PCB180和无饮茶史的联合作用增加结直肠癌的发病风险,OR值为1.51(95%CI：1.03-2.42)。4代谢酶基因多态、有机氯化合物的联合作用与结直肠癌未发现各Ⅰ相代谢酶和Ⅱ相代谢酶基因多态与结直肠癌的易感性有有统计学意义的关联。未发现CYP1A1 MspⅠ基因多态、各有机氯化合物的联合作用与结直肠癌易感性有关。携带CYP1A2 C734A突变型,分别和高暴露的γ-HCH、总HCH、PCB118、PCB180、总PCB、中水平的PCB138的联合作用使结直肠癌的发病风险增加,OR值分别为1.55(95%CI：1.04-2.30)、1.62(95%CI：1.04-2.51)、2.53(95%CI：1.25-5.10)、1.77(95%CI：1.14-2.74)、1.80(95%CI：1.17-2.79)和2.50(95%CI：1.24-5.05)。携带CYP2E1 RsaⅠ突变型,分别和高暴露的γ-HCH、PCB138、PCB180、总PCBs的联合作用使结直肠癌的发病风险增加,OR值分别为1.52(95%CI：1.02-2.25)、1.61(95%CI：1.03-2.52)、1.67(95%CI：1.08-2.59)和1.57(95%CI：1.02-2.42)。携带CYP1B1C1294G突变型、同时分别暴露于高水平的α-HCH、β-HCH、γ-HCH、总HCH、P-p’DDT、P-p’DDE、PCB118、PCB180、总PCBs的个体结直肠癌的发病风险均增加,OR值分别为1.55(95%CI：1.03-2.35)、1.55(95%CI：1.00-2.41)、1.85(95%CI：1.18-2.89)、1.68(95%CI：1.05-2.69)、1.67(95%CI：1.10-2.52)1.57(95%CI：1.03-2.39) 1.74(95%CI：1.11-2.72)、1.75(95%CI：1.10-2.77)和1.89(95%CI：1.19-3.00)。携带GSTM1缺陷型,同时分别暴露于高水平的γ-HCH、PCB28和PCB180结直肠癌的发病风险均增加,OR值分别为1.59(95%CI：1.08-2.36)、1.47(95%CI：1.09-2.19)和1.60(95%CI：1.17-2.31)。未发现GSTT1和有机氯化合物之间存在有统计学意义的联合作用。携带UGT1A6突变型同时暴露于高水平的γ-HCH,结直肠癌的发病风险亦增加,OR值为1.63(95%CI：1.07-2.48)。携带SULT1A1突变型同时分别暴露于中水平的PCB138和PCB180结直肠癌的发病风险也都增加,OR值分别为2.13(95%CI：1.11-4.12)和1.78(95%CI：1.02-3.08)。结论1吸烟可能会增加结直肠癌的发病风险是否有吸烟史以及吸烟量对结直肠癌的发病没有危险性,然而吸烟年限>35年对结直肠癌的发病具有危险效应。2血浆中某些有机氯化合物高暴露可能会增加结直肠癌的发病风险在正常对照人群中,P-p’DDE含量最高,而γ-HCH含量最低。在有机氯化合物和结直肠癌易感性的关系研究中发现,高暴露水平的γ-HCH、P-p’DDE、PCB118以及PCB180分别增加结直肠癌的发病风险。3吸烟和饮酒、有机氯化合物的联合作用可能会增加结直肠癌的发病风险高水平的PCB138和有吸烟史；吸烟年限>35年,分别和高水平的α-HCH、P-p’DDT、P-p’DDE、PCB52、138、180以及总PCBs；吸烟量>20支／天,分别和高水平的总HCH、P-p’DDE、PCB118、180以及总PCBs的联合作用均能增加结直肠癌的发病风险。饮酒年限>31年分别和高水平的PCB52、PCB180的联合作用均能增加结直肠癌的发病风险。4代谢酶基因多态、有机氯化合物的联合作用可能会增加结直肠癌的发病风险CYP1A2突变型,分别和高水平的γ-HCH、PCB118、PCB180；CYP2E1 RsaI突变型,分别和高水平的γ-HCH、PCB138、PCB180；CYP1B1突变型,分别和高水平的各种六六六异构体、P-p’DDT、P-p’DDE、PCB118和PCB180的联合作用均能增加结直肠癌的发病风险。GSTM1缺陷型,分别和高水平的γ-HCH、PCB28和PCB180；SULT1A1突变型,分别和中水平的PCB138、PCB180；以及UGT1A6突变型和高水平的γ-HCH的联合作用也均能增加结直肠癌的发病风险。更多还原

【Abstract】 BackgroundColorectal cancer is one of the most common malignant tumors of digestive tract worldwide. Colorectal cancer is especially common in the North America, Australia and Western Europe, while the incidence tends to be low in Africa and Asia. Recently, with changes of diets and behavior habits, incidence rate of colorectal cancer has been increasing rapidly in China. It is the third prevalent cancer both in males and females with an increasing incidence, of which the estimates in 2005 were 15.0 and 9.7 per 100,000 for males and females, respectively.Colorectal cancer is the result of both effect of environmental factors and genetic susceptibility. Most epidemiological studies have reported that 70 percent of colorectal cancer could be explained by the environmental factors. It is widely accepted that environmental factors, such as diet (high fat and high energy), smoking, drinking, overweight and history of intestinal diseases, play key roles in the susceptibility to colorectal cancer development and progression.Diet is also an important source of exposure to many synthetic organic chemicals used in industry, agriculture, or accidentally released to the environment. Among them, polychlorinated biphenyls (PCBs), the pesticides dichlorodiphenyltrichloroethane (DDT) and lindane [including a-,(3-,y-,8-hexachlorocyclohexane (HCH)] have been classified as "probably" or "possibly" carcinogenic to humans[International Agency for Research on Cancer (IARC)]. Despite reductions in their use and fugitive release, organochlorine compounds remain one of the most important groups of persistent pollutants to which humans are exposed. Many epidemiological studies have reported organochlorine compounds can increase risk of several cancers, including colorectal cancer.Thus, we conducted a population-based case-control study in Jianshan County using serum organochlorine compounds levels as exposure markers, in order to assess the risk of colorectal cancer with exposure to these chemicals, and their potential interactions with environmental factors or genetic polymorphisms of drug-metabolizing enzyme genes in colorectal carcinogenesis.Methods64 693 persons from 10 districts in Jiashan County, Zhejiang Province, who took part in the colorectal cancer screening during 1989 to 1990, consisted of the cohort population of this study.207 eligible patients with confirmed colorectal cancer reported by the cancer registry system were included as cases in our study. As a result of failure in assay detection,202 cases comprised the subjects in this analysis ultimately. Efforts were made to 1:2 individual-match the cases and controls by age with 5-year intervals and gender. A total of 404 controls who did not have a history of cancer were selected randomly from the cohort population during the same period. All individuals were ethnic Han Chinese and residents in Jiashan County.They were interviewed face-to-face through a constructed questionnaire, including demographic characteristics, lifestyles, and disease history, by professionally trained interviewers. In addition, a 5 ml venous blood sample was draw from each individual with the permission. Organochlorine compounds (includingα-,β-,γ-,δ-HCH, DDT, DDE, PCB28,52,101,118,138,153,180) were analyzed by gas chromatography after serum samples were subjected to liquid-phase microextraction.Basic variables of controls and cases were compared by Pearson’s x2tests. Independent-samples t test analysis was used to compare the mean concentrations of organochlorines between cases and controls. Organochlorine concentrations in serum displayed log-normal distributions and therefore, these statistical analyses were performed using the natural logarithm of organochlorines concentrations. In addition, organochlorine compounds concentrations were categorized into three groups using tertiles based on the distributions among control subjects. Conditional logistic regression analysis was performed to calculate the odds ratios (ORs) with 95% confidence intervals (95% CIs). Stratified analyses were used to explore the interaction between organochlorine compounds levels and environmental factors, also between organochlorine compounds levels and genetic polymorphisms. P value of the interactions was evaluated by likelihood test. All P-value were two-sided.Results1. Environment exposures and colorectal cancerThe demographic characteristics did not differ significantly between cases and controls. No association was found between history of smoking, alcohol drinking, tea drinking and the risk of colorectal cancer. However, as adjusted by age, sex, marriage status, education level, occupation, BMI, smokers (>35 years) had a significantly increasing risk of colorectal cancer comparing with non-smokers (OR:2.20,95% CI: 1.13-4.27)2. Organochlorine compounds and colorectal cancerThe serum levels of organochlorine compounds, includingα-HCH,β-HCH,γ-HCH,δ-HCH, P-p’DDT, P-p’DDE, PCB28, PCB 52, PCB101, PCB118, PCB138, PCB153, PCB180, also did not differ significantly between cases and controls. However, when stratified analyse was used, we found increased risk of colorectal cancer was associated with high levels of y-HCH (>4.03 ng/g), P-p’DDE (>1157.61ng/g), PCB118 (>36.79ng/g), PCB180 (>8.09ng/g) and total PCBs (>341.53ng/g). The ORs were 1.36 (95%CI:1.04-1.77),1.31 (95%CI:1.01-1.71),1.36 (95%CI:1.05-1.67),1.45 (95%CI: 1.06-1.99) and 1.35 (95%CI:1.00-1.82) in each highest group, respectively.3. Interaction of lifestyle factors and organochlorine compoundsIndividuals with a history of smoking and exposed of high level of PCB138 had a significantly increasing risk of colorectal cancer (OR:1.96,95% CI:1.19-3.23). Interactions existed between smokers (>35 years) and high levels of a-HCH, P-p’DDT, P-p’DDE, PCB52, PCB 138, PCB 180 and total PCBs, respectively. The ORs were 2.00 (95% CI:1.09-3.65),1.82 (95% CI:1.03-3.22),1.95 (95% CI:1.11-3.43),1.92 (95% CI: 1.08-3.44),1.93 (95% CI:1.02-3.66),2.61 (95% CI:1.33-5.11) and 2.44 (95% CI: 1.35-4.41). Interactions also existed between heavy smokers (>20 cigarette per day) and high levels of total HCH, P-p’DDE, PCB118, PCB180 and total PCBs, and the ORs were 1.82 (95%CI:1.00-3.32),2.03 (95%CI:1.14-3.60),2.08 (95%CI:1.13-3.83),2.72 (95%CI:1.41-5.23) and 1.92(95%CI:1.08-3.39), respectively.No significant associations were found between alcohol drinking and organochlorine compounds with the risk of colorectal cancer. However, the risk of colorectal cancer increased when alcohol drinking (>31 years), combined with high levels of PCB52 and PCB180, respectively. The ORs were 1.94 (95%CI:1.00-3.78) and 1.77(95%CI:1.05-3.28).What’s more, individuals with a history of tea drinking and exposed of low levels of P-p’DDT decreased colorectal cancer risk (OR:0.59,95%CI:0.37-0.94). Individuals without a history of tea drinking and exposed of high levels of PCB180 increased colorectal cancer risk (OR:1.51,95%CI:1.03-2.42).4. Interactions between genetic polymorphisms and organochlorine compoundsNo significant association was found between phaseⅠenzymes, phaseⅡenzymes gene polymorphisms and colorectal cancer risk.The risk of colorectal cancer increased significantly when individuals with CYP1A2 C734A variant, combined with high levels ofγ-HCH, total HCH, PCB118, PCB180, total PCBs, middle level of PCB138, respectively. The ORs were 1.55 (95%CI: 1.04-2.30),1.62 (95%CI:1.04-2.51),2.53 (95%CI:1.25-5.10),1.77 (95%CI:1.14-2.74), 1.80 (95%CI:1.17-2.79) and 2.50 (95%CI:1.24-5.05). Also, the risk increased when individuals with CYP2E1 Rsa I variant, combined with y-HCH, PCB138, PCB180 and total PCBs, respectively. The ORs were 1.52 (95%CI:1.02-2.25),1.61 (95%CI: 1.03-2.52),1.67 (95%CI:1.08-2.59) and 1.57 (95%CI:1.02-2.42). In addition, individuals with CYP1B1 C1294G variant, together with high levels of a-HCH,β-HCH,γ-HCH, total HCH, P-p’DDT, P-p’DDE, PCB118, PCB180 and total PCBs, ryespectively, elevated the risk of colorectal cancer. The ORs were 1.55 (95%CI: 1.03-2.35),1.55 (95%CI:1.00-2.41),1.85 (95%CI:1.18-2.89),1.68 (95%CI:1.05-2.69), 1.67 (95%CI:1.10-2.52),1.57 (95%CI:1.03-2.39),1.74 (95%CI:1.11-2.72),1.75 (95%CI:1.10-2.77) and 1.89 (95%CI:1.19-3.00).GSTM1-deleted genotype associated with high levels y-HCH, PCB28, PCB180, respectively, had a significantly increasing risk of colorectal cancer. The ORs were 1.59 (95%CI:1.08-2.36),1.47 (95%CI:1.09-2.19) and 1.60 (95%CI:1.17-2.31). Additionally, UGT1A6 variant associated with high level of y-HCH, SULT1A1 variant associated with middle level of PCB138, SULT1A1 variant associated with middle level of PCB180, all elevated the risk of colorectal cancer. The ORs were 1.63 (95%C7: 1.07-2.48),2.13 (95%CI:1.11-4.12) and 1.78 (95%CI:1.02-3.08), respectively.ConclusionsConclusions are drawn from the results of the population-based case-control study as followed:1. Smoking (>35 years) may contribute to the colorectal cancer predisposition.2. High levels of y-HCH, P-p’DDE, PCB118 and PCB180 may have effect on colorectal cancer.3. Interactions of smoking, especially its duration (>35 years) and amount (>20 cigarette per day), and high levels of organochlorine compounds may have influence on colorectal cancer. In addition, alcohol drinking may have similar effect.4. Phase I, phase II enzymes genetic polymorphisms may have a modifying effect on the risk of colorectal cancer associated with exposures to organochlorine compounds.更多还原