BACH1 Ser919Pro variant and breast cancer risk

Breast cancer patients belonging to 43 breast cancer families with at least three breast or ovarian cancer cases in 1^st or 2^nd degree relatives and with no detectable BRCA1/2 mutations were included in the initial mutation analysis. Recruitment of the families through the Department of Oncology, Helsinki University Central Hospital, Finland as well as verification of the cancer diagnoses and exclusion of the BRCA1 and BRCA2 mutations have been previously described [19‐21].

The BACH1 variant Ser919Pro was analyzed in a large series of unselected breast cancer patients and healthy controls. The 888 unselected breast cancer patients were collected at the Helsinki University Central Hospital, Finland, during April 1997-March 1998 [22] and January-June 2000 [23], and cover 79% of all consecutive, newly diagnosed breast cancer cases during the collection period. DNA samples from altogether 736 healthy females collected at the same geographical region of Southern Finland were studied as healthy population controls. As the variant was associated with an increased breast cancer risk by the age of 50 years [14] the study cohort as well as the population controls were subgrouped according to the menopausal status (age 50 years was chosen as a surrogate for menopause, and patients with cancer diagnosis at < 50 years were considered premenopausal and ≥50 years as postmenopausal). Breast tumor characteristics (tumor histology, size, and grade; nodal and distal metastasis; estrogen and progesterone receptor status) were available from all patients. Additionally, a Val193Ile variant that was found in only one family in the initial mutation analysis was further genotyped in randomly selected series of 346 familial breast cancer patients and in 183 healthy population controls. All mutation analyses have been performed on DNA samples extracted from peripheral blood.

The study was performed with informed consent from the patients and under appropriate research permissions from the Ethics Committees of the Departments of Obstetrics and Gynecology, and Oncology, Helsinki University Central Hospital, Finland, as well as Ministry of Social Affairs and Health in Finland.

The coding region and exon-intron boundaries of BACH1 were screened using conformation sensitive gel electrophoresis (CSGE) with modifications previously described [24]. Samples with aberrant CSGE profiles were reamplified, and the nucleotide changes determined by direct sequencing. Primers and PCR conditions used in the mutation analysis are presented in table 1. The frequency of the Ser919Pro alteration was determined by Amplifluor™ fluorescent genotyping (K-Biosciences, Cambridge, UK). The genotyping for nt c.2755 (codon 919) was successful in 866/888 (97.5%) breast cancer patient samples and in 731/736 (99.3%) healthy control samples. The Val193Ile (c.577G>A) variant was genotyped in 346 (successful in 336, 97.1%) familial breast cancer samples and in 183 (successful in 167, 91.3%) control samples, respectively, by Amplifluor™ fluorescent genotyping as well.

Possible associations between the BACH1 Ser919Pro and breast cancer risk, as well as the variant and clinico-pathologic features of the tumors, were tested by univariate analysis. Independent variables were compared with the chi-square test. The mean age at breast cancer diagnoses between the carriers and non-carriers was compared by one-way ANOVA. All p-values were 2-sided, and p-value < 0.01 was considered statistically significant as suggested by Houlston and Peto [25]. All statistical analyses were carried out in the SPSS software (version 12.0 for Windows, SPSS, Chicago, IL, USA).

Altogether six germ line BACH1 variants were observed in the initial mutation analysis (Table 2). Four of the changes were in the coding region, two of these were missense and two silent substitutions. The missense changes were analyzed further for their possible association with breast cancer risk. The silent alterations have been suggested as neutral polymorphisms by the previous studies [2, 26, 27], and were not studied here further. The one missense substitution, Val193Ile, was seen in only one family. In addition to the proband diagnosed with breast cancer at the age of 54 years, the variant was also seen in her healthy father and healthy brother, father's sister diagnosed with skin cancer at age 85 years, and the aunts' three children diagnosed with breast cancer at 63 years, ovarian cancer at 59 years, and skin cancer at 64 years, respectively. The grandmother of the proband, whose carrier status is unknown, has been diagnosed with breast cancer at age 74 years. The variant was not observed among 336 familial breast cancer patients or in 167 healthy population controls. It has previously been observed in 3/200 (1.5%) healthy controls and classified as a rare polymorphism [2]. The residue resides in close proximity to the ATP/GTP binding site, and the possible functional significance of this rare variant to BACH1 protein function remains to be determined.

The other observed missense substitution, a common polymorphism Ser919Pro, was associated with an elevated breast cancer risk during the course of this study [14]. In that study the relative cumulative risk by the age 50 years was 4.5 for the female first degree relatives of Pro/Ser heterozygotes (95% CI 0.8–12.2; p = 0.096) and up to 6.9 (95% CI 1.6–29.3; p = 0.018) for the first degree relatives of Ser/Ser homozygotes, when compared to Pro/Pro homozygotes. However, no significant association was seen when the analysis was extended to age 70 years (OR 1.3, 95%CI 0.8–2.8, p = 0.220) [14]. Here we found no association of the variant with breast cancer risk. The odds ratios for the whole study cohort as well as for the pre- and postmenopausal patient groups were close to one both for the Pro/Ser heterozygotes and for the Ser/Ser homozygotes when compared to Pro/Pro homozygotes (Table 3). The alteration did not associate with breast cancer family history (data not shown). No association of the variant with any of the breast tumor characteristics (Table 4) or survival (data not shown) was seen either, and also the age at breast cancer diagnosis was similar in all genotype carrier groups (56.9 years for the Pro/Pro homozygotes, 56.8 years for the heterozygotes, and 56.1 years for the Ser/Ser homozygotes, respectively; p = 0.705). Our data suggest that the BACH1 Ser919Pro alteration is not a breast cancer predisposition allele in our study population although a very low risk effect cannot be excluded. A joint effect on breast cancer risk by the Ser919Pro variant and other epidemiological risk factors may also be possible.

Our data, together with previous studies, suggest that germ line mutations in BACH1 do not substantially contribute to the remaining proportion of the familial aggregation of breast cancer outside the high-penetrance genes BRCA1 and BRCA2 [2, 26‐28]. In the future, it will be interesting to see whether the heterozygous BACH1 mutation carriers in FA families have an excess risk of developing breast or other type of cancer. So far, heterozygous mutations of the FA genes, with the exception of BRCA2, have been found to be extremely rare, or nonexistent, in breast cancer families [29], and the studied polymorphisms have not been found to confer an increased risk for breast cancer [[30], this study].