Case-Control Study On The Associations Of Genetic Polymorphisms Of CYP1A1 MspⅠ, ER PvuⅡ And ER XbaⅠ With The Risk Of Breast Cancer

【Background】 Over the lastest 20 years, the incidence rate of breast cancer has been increasing rapidly in most areas such as Beijing, Tianjin and Shanghai of China. It is becoming the major malignant tumor treatening health and life of women. Based on the current knowledge, only about 30~40% of breast cancer cases could be explained by the well-established risk factors. It is urgent to explore the new risk factors for the breast cancer in term of the distinct perspectives. Many studies showed that the estrogen itself and other estrogen-related factors may be associated with the breast cancer,and the genetic polymorphisms of estrogen-metabolizing gene and estrogen receptor gene were thought to be the risk factors in the process of the breast cancer initiating and developing by the estrogen. Hence, we hypothesized that there might have the gene-gene and gene-environment joint risk effects on the breast cancer among the polymorphisms of the CYP1A1 MspI, ER PvuII and ER XbaI, and those polymorphisms and some environment exposures regarding the estrogen. 【Objective】 We are sought to examine whether the 3 genetic polymorphisms could increase the breast cancer risk of Chinese women, and to further estimate gene-gene and gene-environment joint risk effects on the breast cancer. 【Method】 We extracted subjects from a population-based case-control study 'Induced abortion and breast cancer risk', which was carried out in Shanghai between Jan 1, 1999 and Apr 30, 2001. Using the collected buccal cell samples, we conducted the PCR-based RFLP assays to determine susceptibility of the breast cancer from the role of genetic polymorphisms in related to estrogen. Taking into consideration of the success rate of experiment, five hundreds and eighty buccal cell samples, collected between Dec 6, 2000 and Apr 30, 2001, were used to make the DNA extraction and PCR-RFLP experiment. Five hundreds and thirty seven subjects, namely 259 cases and 278 controls, entered into final analysis. Our experiment work spanned from Mar 1, 2002 to Jan 1, 2004. Odds ratio (OR) and 95% confience interval (CI) were used to estimate the association between risk factors and breast cancer. Qulity control measures, such as the RFLP graph comparasion between buccal cells and blood samples, repeated measure RFLP results in 10% samples, and double data entry were used. The statistical methods included chi-square test, Hardy-Weinberg balance test, unconditional logistic regression and Bruzzi's PAR% estimation. SAS 8.1 software was used to anlyze the data. 【Results】 Premenopausal status, being younger at menarche, total number of menstrual cycles, number of menstrual cycles prior to FFTP (first full-term pregnancy), family history of breast cancer, benign breast disease history, higher exposure to photographs/X-rays before 15 years old, pork meat and preserved foods intakes were associated with an increased risk of breast cancer. Body height was the risk factor of breast cancer only in the postmenopausal women. Physical exercises and being younger at first full-term pregnancy could decrease the risk of breast cancer. No associations were found between oral contraceptives (OCs), soybean foods and breast cancer risk. However, the use of sexual hormone showed a positive trend of increasing the risk of breast cancer but did not reach a statistically significant level. The frequency of wild-type, heterozygote and homozygote for the CYP1A1MspI, ER PvuII and ER XbaI in control women of Shanghai were 47.8%, 40.7%, 11.5%;15.7%, 45.3%, 39.0% and 7.6%, 31.5%, 60.9%, respectively. We found that the polymorphism of the estrogen metabolizing gene, CYP1A1 MspI was associated with an elevated risk of breast cancer. Compared with the wild genotype (m1/m1), the adjusted OR for the m1/m2 and m2/m2 were 1.98(95%CI: 1.33-2.98) and 2.17(95%CI: 1.22-3.88), respectively, indicating an increasing trend of the risk with the number of mutation allele (trend test P<0.01). The ER PvuII, XbaI polymorphisms probably had the weak associations with b...