Association of CYP1B1 Haplotypes and Breast Cancer Risk in Caucasian Women

Cytochrome P450, family 1, subfamily B, polypeptide 1 (CYP1B1) is a key member of the cytochrome P450 (CYP) superfamily of enzymes (1). The encoded estrogen-metabolizing protein catalyzes the hydroxylation of estradiol (2), an estrogenic hormone that plays an important role in the development and progression of hormone-related cancers such as breast cancer. Overexpression of CYP1B1 may cause accumulation of estrogen metabolites that directly or indirectly damage DNA (3). A recent meta-analysis found no association between the CYP1B1 Leu432Val polymorphism (rs1056836) and risk of breast cancer among Caucasian women (4). Associations with breast cancer have been examined for other putative functional polymorphisms in CYP1B1 (5, 6). One previous analysis in Polish women showed no significant haplotype effects among eight CYP1B1 tagging single nucleotide polymorphism SNPs (7). We undertook a comprehensive evaluation of common genetic variations in CYP1B1 with breast cancer risk in a large U.S. population–based case-control study.

The study population has been previously described (8). Briefly, the subjects enrolled in this population-based study were all English-speaking females, residing in Massachusetts, New Hampshire, and Wisconsin. Case women had a recent diagnosis of invasive breast cancer, were ages 20 to 69 y, and were identified through state-wide cancer registries. Controls were randomly selected from lists of licensed drivers (if <65 y) and from a roster of Medicare beneficiaries (if 65-69 y) in each state. DNA samples were collected through the mail from 70% of interviewed cases and 61% of controls using an oral rinse protocol (9).

We genotyped the same haplotype-tagging SNPs in CYP1B1 that were examined in the study by Gaudet et al. (7) of Polish women. The SNP identification process, based on strategies used in the Breast and Prostate Cohort Consortium (10), was described in the earlier report (7). In brief, SNPs were identified by a search of public databases and by resequencing exonic regions of the gene in a multiethnic panel of 190 patients with advanced breast or prostate cancer enrolled in the Multiethnic Cohort. This sample size was chosen to provide >85% power to discover variants with a minor allele frequency of >5% present in a single population group. Using the TagSNP program (11), a minimum set of 8 SNPs was selected to capture common haplotype diversity in CYP1B1 among Causcasian women specifying a pairwise r² of >0.8. We genotyped all the eight haplotype-tagging SNPs, including one SNP (rs1800440), which causes an amino acid change (Asn453Ser).

Isolation and preparation of the genomic DNA samples have been previously described (9). Genotypes were evaluated using validated Taqman or MGM Eclipse assays as described in detail at the snp500cancer Website⁹

We restricted the analysis to Caucasian women who comprised 98% of the study population (1,655 invasive cases and 1,470 controls). Genotypes for 7 of the 8 SNPs were consistent with Hardy-Weinberg equilibrium (P ≥ 0.05) based on an exact test (13); 1 SNP (rs162557) had a borderline departure from Hardy-Weinberg equilibrium (P = 0.03).

For each individual SNP, multivariable unconditional logistic regression was used to obtain age- and state-adjusted odds ratio (OR) estimates and 95% confidence intervals (CIs). Tests for linear trend in breast cancer risk for each individual SNP were conducted by including an ordinal coding of the genotypes of the SNP in the logistic regression models. Haplotype frequencies and effects were examined using the statistical program Haplo.stats (14). The function Haplo.score was used to estimate haplotypes and assess differences in haplotype frequencies between cases and controls. A global score test was used to evaluate the overall significance, with adjustment for age group and residence. Effects of individual haplotypes were also examined by comparing the breast cancer risk associated with each inferred haplotype to the risk associated with the highest estimated frequency haplotype. The estimated ORs and 95% CIs were obtained using the function Haplo.glm. Rare haplotypes with frequencies of <1% were pooled into a single category to reduce the burden of sparse table cells.

The results of the association analysis for individual CYP1B1 SNPs are summarized in Table 1. Results for one uncommon SNP (rs9341266) had a significant test for trend with increasing number of minor alleles (P_trend = 0.021). Results were essentially unchanged after adjustment for established breast cancer risk factors.

One LD block was identified in our study population (data not shown). Eight CYP1B1 haplotypes were identified with a frequency >1% (Table 2). One haplotype comprising 3.1% of controls (CCTAACTA) was associated with a modestly elevated relative risk (OR, 1.40; 95% CI, 1.05-1.87). One other haplotype found in 1% of controls (CTCAATTG) was associated with an approximate doubling in the relative risk (OR, 2.00; 95% CI, 1.23-3.27) when compared with the most common reference haplotype. A marginally significant haplotype effect was found when comparing breast cancer cases with the controls (P_global = 0.015).

We considered results for individual SNPs according to menopausal status, body mass index, and use of hormone replacement therapy, and no differences were noted (data not shown).

To our knowledge, this is one of the first studies that used haplotype-tagging SNPs to examine the influence of CYP1B1 haplotypes on breast cancer risk. Using this approach, we found no evidence that Caucasian women with any common haplotype in CYP1B1 have a substantially altered risk of developing invasive breast cancer. Despite a relatively large sample size, the study had limited power to examine uncommon haplotypes and interactions with potential effect modifiers, and we lacked information on steroid receptor status. Results are in line with one previous study that included an identical panel of haplotype-tagging SNPs (7), which found no significant associations with breast cancer among inferred CYP1B1 haplotypes. A genome-wide association study based on the Cancer Genetic Markers of Susceptibility Project also did not identify breast cancer associations with markers at chromosome 2p22-p21 where CYP1B1 resides (15). The study identified associations of individual CYP1B1 variants with breast cancer risk and does not rule out a minor haplotype effect in Caucasian women.

No potential conflicts of interest were disclosed.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.