Genetic Polymorphisms in the Estrogen Receptor α Gene and Risk of Breast Cancer: Results from the Shanghai Breast Cancer Study¹

Estrogen influences the growth, differentiation, and function of many target tissues, including the breast, uterus, vagina, ovary, testis, epididymis, and prostate (1). The biological effect of estrogens such as stimulating growth and differentiation of normal mammary tissue is mediated primarily through high-affinity binding to ERs³ (2). ERs are nuclear receptor proteins that have an estrogen binding domain and a DNA binding domain (3, 4). There are two types of ERs, ER-α and ER-β. The ER-α gene is localized on chromosome 6q25.1 (5), and the ER-β gene is localized on chromosome 14q22-24 (6). Among the steroid receptors, ER-α and the ER-regulated PR are of special interest because their protein levels are elevated in premalignant and malignant breast cells (7). Both ER and PR have been demonstrated to be significant prognostic factors for breast cancer (8). Consequently, inhibition of the ER-α has become one of the major strategies for the prevention and treatment of breast cancer (9).

The association of genetic polymorphisms in the ER-α gene and the risk of diseases, including breast cancer, have been the subject of increasing interest. Several DNA sequence variations in the ER-α gene have been reported. Two case-case studies showed that PvuII polymorphism was related to a younger age at breast cancer diagnosis (10, 11). Two case-control studies reported that XbaI, not PvuII, polymorphism was related to breast cancer risk (12, 13). The sample sizes from these studies, however, were small in general, and potential selection bias in hospital-based case-control study was a concern.

In this article, we report results from the Shanghai Breast Cancer Study, a large-scale population-based case-control study that examined the association of ER-α gene PvuII and XbaI polymorphisms with the risk of breast cancer. The associations of these polymorphisms with ER/PR status among breast cancer cases were also evaluated.

Case patients and control subjects in this study were participants of the Shanghai Breast Cancer Study, a population-based case-control study (14, 15, 16). This study included 1459 women who were between the ages of 25–64 years and diagnosed with breast cancer from August 1996 through March 1998, as well as 1556 age frequency-matched control women. The study protocol was approved by committees of relevant institutions for the use of human subjects in research. All case patients and control subjects were permanent residents of urban Shanghai who had no histories of breast cancer. Through a rapid case ascertainment system supplemented by the population-based Shanghai Tumor Registry, 1602 eligible case patients with breast cancer were identified during the study period, and in-person interviews were completed for 1459 (91%) of them. The major reasons for nonparticipation were refusal (109 case patients, 6.8%), death before the interview (17 case patients, 1.1%), and the inability to locate (17 case patients, 1.1%). Cancer diagnoses for all patients were confirmed by two senior study pathologists through a review of tumor slides. Detailed information on cancer diagnosis and treatment, including ER and PR status, was abstracted from medical charts. The information on ER/PR status was obtained from 956 of the 1459 breast cancer cases. Of those, 52.7% were ER+/PR+, 11.2%, ER+/PR−, 10.6%, ER−/PR+, and 25.5%, ER−/PR−.

Control subjects were randomly selected from the female general population and were frequency matched to case patients by age (5-year intervals). The number of control subjects in each age-specific stratum was determined in advance according to the most recent data on the age distributions of the breast cancer patients available from the Shanghai Tumor Registry. The Shanghai Resident Registry, which keeps registry cards for all adult residents in urban Shanghai, was used to randomly select control subjects. For each age-predetermined control subject, a registry card identifying a potential control subject in the same 5-year age group was randomly selected. Only the women who lived at the address during the study period were considered to be eligible for the study. In-person interviews were completed for 1556 (90%) of the 1724 eligible control subjects identified. Excluded from the study were 168 potential control subjects because of refusal (n = 166; 9.6%) and death or a prior cancer diagnosis (n = 2; 0.1%).

A structured questionnaire was used to elicit detailed information on demographic factors, menstrual and reproductive histories, hormone use, dietary habits, prior disease history, physical activity, tobacco and alcohol use, weight, and family history of cancer. All participants were also measured for their current weight, circumferences of the waist and hip, and heights while sitting and standing. Blood samples were obtained from 1193 (82%) case patients and 1310 (84%) control subjects who completed the in-person interviews. These samples were used for the genotyping assays in this study.

Genomic DNA was extracted from buffy coat fractions using the Puregene DNA Isolation kit (Gentra Systems, Minneapolis, MN) following the manufacturer’s protocol. DNA concentration was measured by PicoGreen dsDNA Quantitation kit (Molecular Probes, Eugene, OR). Ten ng of genomic DNA were used for each PCR.

ER-α genotypes were determined with a PCR-RFLP method reported earlier (10, 17) with some modification. The primers for analysis were: 5′-CTGCCACCCTATCTGTATCTTTTCCTATTCTCC-3′ (forward) and 5′-TCTTTCTCTGCCACCCTGGCGTCGATTATCTGA-3′ (reverse). These primers generated a 1.3-kb fragment. The PCR was performed in a PTC-200 Peltier Thermal Cycler (MJ Research, Inc., Waltham, MA). Each 35 μl of PCR mixture contained 10 ng of DNA, 1× PCR buffer [50 mm KCl, 10 mm Tris-HCl (pH 9.0)], 2.5 mm MgCl₂, 0.16 mm deoxynucleoside triphosphate, 0.5 μm of each primer, and 1.5 unit of TaqDNA polymerase. The reaction mixture was initially denatured at 94°C for 3 min, followed by 36 cycles of 94°C for 45 s, 61°C for 45 s, and 72°C for 2 min. The PCR was completed by a final extension cycle at 72°C for 7 min. The product contains a part of intron 1 and exon 2 of the ER-α gene.

The PCR products were digested by the PvuII and XbaI restriction endonucleases, respectively. The DNA fragments were then separated using 1.5% agarose gel and detected by ethidium bromide staining. PP and XX, signifying the absence of restriction sites, gave one 1.3-kb fragment. pp, signifying the presence of PvuII restriction sites on both alleles, was digested into two fragments (0.85 and 0.45 kb). The xx genotype was revealed by XbaI digestion into two fragments (0.9 and 0.4 kb).

The laboratory staff was blind to the identity of the subject. QC samples were included in genotyping assays. Each 96-well plate contains one water, two Centre d’Etude du Polymorphisme Humain (CEPH) 1347-02 DNA, two blinded QC DNA, and two unblinded QC DNA samples. The blinded and unblended QC samples were taken from the second tube of study samples included in the study. The ER-α genotypes determined for the QC samples were in complete agreement with the genotypes determined for the study samples.

Genotyping data were obtained from 1069 (89.6%) cases and 1166 (89.0%) controls who have blood samples, which represent 67% (1069 of 1602) of eligible case patients and 67% (1166 of 1724) of eligible control subjects. The major reasons for incomplete genotyping were insufficient DNA and unsuccessful PCR amplification.

χ² statistics were used to evaluate case-control difference in the distribution of allele types and genotypes. To accommodate the age frequency-matched study design, we used logistic regression models conditioned on age to estimate ORs and 95% CIs to measure the strength of the association between ER-α gene polymorphisms and breast cancer risk (18, 19). Although unconditional logistic regression could also be used by including age variable in the model, conditional logistic regression is more direct and intuitive in the control for the effect of matched variables and is more parsimonious because it does not include matched variables in the model. To better assess CIs for estimates based on the small reference group, the FAR CIs were analyzed (20). Analyses stratified by age and menopausal status were conducted to check the homogeneity of the association. Additional analyses stratifying by indicators of endogenous estrogen exposure and lifestyle factors were conducted to evaluate the potential modifying effects of these variables on the association between ER-α genotypes and breast cancer risk. All statistical tests were two-sided.

The distributions of selected demographic characteristics and major risk factors for breast cancer are shown in Table 1. These data are presented separately for subjects with genotyping data and those included in the whole study to evaluate the consistency of the distribution of these factors between these two groups. There was no appreciable difference between cases included in the genotyping study and the whole study. The same was true for the two control groups. Breast cancer cases and controls were comparable in age and education level. Although not all statistically significant, elevated risks of breast cancer were observed for all major risk factors that have been reported from studies conducted elsewhere (21). The associations of these risk factors with breast cancer were similar in the subjects with genotyping data and subjects in the whole study (Table 1).

Allele frequencies of ER-α gene PvuII and XbaI polymorphisms in the control group were similar to those reported previously from other studies conducted in Asian populations (13, 17). The P allele was slightly more prevalent among controls (39.7%) than cases (37.1%) in the PvuII polymorphism (P = 0.07). Approximately 16.3% of controls and 12.9% of cases were homozygous, and 46.8% of controls and 48.3% of cases were heterozygous for this allele (P = 0.08). Age-adjusted ORs for genotypes Pp and pp were 1.3 (95% CI, 1.0–1.7) and 1.4 (95% CI, 1.1–1.8), respectively, comparing to genotype PP (Table 2). There was no appreciable difference in allele frequency or genotype of the XbaI polymorphism between controls and cases (Table 2). When the two ER-α polymorphisms were analyzed jointly, no consistent synergistic effect was noted. The genotype combinations, including one or two palleles, were, in general, associated with an elevated risk, and the highest OR was observed for the genotype with pp and Xx combination (OR = 1.7, 95% CI, 1.0–3.1). The ORs were essentially unchanged after additional adjustment of physical activity, BMI, WHR, menarche, live girth, age at first birth, and breast cancer family history. Also presented in Table 2 are FAR. The range of these risk estimates are narrower than the 95% CI estimated for point estimate of risk.

Table 3 showed the associations of ER-α genotypes with breast cancer risk by age and menopausal status. The PvuII polymorphism was associated with elevated breast cancer risk in all strata defined by age and menopausal status. For the XbaI polymorphism, however, the elevated risks were mainly confined to older (>45 years) or postmenopausal women.

Additional analyses were conducted to evaluate the association of the ER-α gene polymorphisms and risk of breast cancer by indicators of endogenous estrogen exposure (Table 4). These indicators included years of menstruation, a prior history of breast fibroadenoma, BMI, and WHR. The risk associated with the p allele of the PvuII polymorphism was elevated in all strata, and no apparent modifying effect of endogenous estrogen exposure indicators was observed. We also evaluated the association of the ER-α gene polymorphisms and risk of breast cancer by lifestyle factors. The proportions of smokers, drinkers, and hormone replacement therapy user were very small, making it difficult to evaluate their modifying effects. No apparent modifying effects were found for physical activity, calorie intake, fat intake, protein intake, and carbohydrate intake (Table 4). Similarly, no modifying effect of endogenous estrogen exposure indicators and lifestyle factors on the association between XbaI polymorphism and breast cancer risk was found.

We also evaluated the association of ER-α gene PvuII and XbaI polymorphisms with ER/PR status among breast cancer cases. No apparent association was observed (Table 5).

The association of ER-α genetic polymorphisms with breast cancer risk attracts much attention because ER functions as a hormone-dependent transcriptional regulator, which, in turn, plays a significant role in the development of breast cancer (1, 22). Several ER-α gene polymorphisms have been reported, among which PvuII and XbaI polymorphisms are the most studied. Several diseases, including breast cancer (10, 11, 12, 13), endometrial cancer (23), Alzheimer’s disease (24), obesity (25), multiple sclerosis (26), endometriosis (27), adenomyosis (28), leiomyomas (27, 28), and bone mineral density (17, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34), have been evaluated for possible linkage with PvuII and XbaI polymorphisms.

Both PvuII and XbaI polymorphisms are located in intron 1 of the ER-α gene and are 50 bp apart (10, 35). Parl et al. (11) found that the ppgenotype of PvuII was related to a younger age at breast cancer diagnosis. Yaich et al. (10) examined the PvuII polymorphism in the tumor tissue of 257 primary breast cancer patients and 140 peripheral blood DNA samples from women without breast cancer. Breast cancer patients with a ppgenotype were significantly younger than women with PP or Pp genotype at the time of cancer diagnosis. However, this finding was found in only one of the two study hospitals (10). In a study of 360 breast cancer patients from a hospital in Norway and 672 convenient controls, Andersen et al. (12) found that allele frequencies of the PvuII polymorphism did not differ between cases and controls. The frequency of the x allele of the XbaI polymorphism among breast cancer patients, however, was 1.4 times of that for controls (95% CI, 1.0–1.9; Ref. 12). Among the breast cancer patients, there was an association of borderline significance between the XbaI restriction site and older age at onset (12). In a hospital-based case-control study (201 cases and 201 controls) conducted in South Korea, Shin et al. (13) reported that OR associated with the xx genotypes of XbaI was 2.38 (95% CI, 1.58–3.58) compared with women with XX genotype. In our large-scale population-based case-control study, we found that the polymorphism at the PvuII restriction site (pallele) was associated with an elevated risk of breast cancer. The associations were, in general, weak with the highest OR of 1.5. The XbaI polymorphism was associated with a nonsignificant increased risk for breast cancer only among older or postmenopausal women. The differences in study design (hospital based versus population based) and study population (Caucasian versus Chinese) may have contributed to the inconsistency between findings from previous and current studies.

How breast cancer risk is affected by the intronic PvuII polymorphism of the ER-α gene remains unclear. Possible explanations include: (a) the intronic polymorphism may be in linkage disequilibrium with exon alteration, which affects ER protein function (17); (b) the PvuII polymorphism in the ER-α gene may be linked with the alteration of another unidentified gene adjacent to the ER-α gene, which increases breast cancer risk (17); (c) intronic changes in gene sequence may have an impact on the expression of other genes by influencing the transcription and/or stability of mRNA of those genes (36, 37); (d) and some introns contain regulatory sequences such as enhancers, which affect the levels of expression through transcriptional regulation (38, 39). For example, it has been shown that a polymorphism within the first intron of Drosophila β-3-tubulin gene has a significant effect on the level of the protein synthesis (39). The PvuII polymorphism has been previously reported to be associated with ER expression in a study of 188 breast cancer patients (40). Along with two later studies (10, 12), we did not find this association in our study. It has also been indicated that the PvuII restriction site (p allele) is associated with PR negativity (12). Additional functional analyses of this polymorphism are needed to better understand how the PvuII polymorphism is involved in breast cancer development.

Other genetic polymorphisms of ER-α gene may also be associated with breast cancer risk. In a hospital-based study, a statistically significant association was found between the polymorphism in ER-α codon 325 [CCC→CCG (Pro)] and a self-reported family history of breast cancer (2). However, no association was found between ER-α gene polymorphism in codon 325 and the risk of breast cancer in a population-based case-control study of breast cancer among younger women (<40 years) conducted in Australia (41). Recently, Kang et al. (42) reported that this silent mutation in exon 4 was associated with tumor PR and ER expression. Additional studies are needed to investigate the association of these polymorphisms, as well as their interaction with PvuII and XbaI polymorphisms. In addition, growth factors and their signaling molecules are important for breast cancer growth and progression. There is considerable cross-talk between ER and growth factors pathways (43). For example, the PI3K pathway mediates cell survival and proliferation signals coming from growth factors such as insulin, the insulin-like growth factors, and epidermal growth factor family (44). ER directly interacts with PI3K in the cell membrane to activate it (45). The PI3K/AKT pathway can also directly modulate the ER by phosphorylating ER at serine 167 (46). The mitogen-activated protein kinase and stress response pathways also interact with ER (43). It would be useful to study the interaction of the ER-α gene with genes involved in these pathways.

The current study has many strengths. First, the high participation rate and population based study design substantially reduced selection bias. Second, Chinese women living in Shanghai are relatively homogeneous in ethnic backgrounds; over 98% of them are classified into a single ethnic group (Han Chinese). Therefore, the potential confounding effect by ethnicity is not a major concern in our study. Third, the extensive information on lifestyle factors allowed a comprehensive evaluation of their interaction with the genetic polymorphisms.

In summary, in this population-based case-control study, we found that PvuII polymorphism in the ER-α gene was associated with breast cancer risk. Additional studies are needed to understand the nature of the association.