RASSF1A Polymorphism A133S Is Associated with Early Onset Breast Cancer in BRCA1/2 Mutation Carriers

Five percent to 10% of all breast and ovarian cancers are due to a genetic predisposition (1). The most common form of familial breast and ovarian cancers are associated with germ line mutations in BRCA1 or BRCA2 (2, 3). Inherited mutations in other genes including CHEK2, PTEN, TP53, ATM, NBS1, RAD50, BRIP1, and PALB2, although rare, have been associated with an increased risk of developing breast cancer (4, 5). Together, mutations in these genes account for ∼25% of familial cases of breast cancer, leaving the majority of cases without an explanation. In addition, there is substantial variability in the incidence of breast cancer among mutation carriers, and the age of onset also varies widely (6). These observations suggest that unknown genetic factors contribute to the development of familial breast cancer.

We and others have shown that RASSF1A is a tumor suppressor gene located on chromosome 3p21.3, an area with frequent loss of heterozygosity (LOH) in lung and breast cancers (7). RASSF1A is expressed in all nonmalignant epithelial cells, but not in more than half of breast cancers due to hypermethylation in the promoter region (8, 9). Additionally, we have shown that RASSF1A plays an important role in cell cycle control by preventing cyclin D1 protein accumulation (10), whereas others have shown that RASSF1A plays important roles in mitosis, apoptosis, and microtubule stabilization (11–14).

Mutations in RASSF1A are rare. However, several polymorphisms have been identified. One polymorphism, Ala133Ser (A133S), is located close to a putative ATM substrate site (S131) in the microtubule-binding domain. We have shown that ectopic expression of wild-type RASSF1A inhibits the accumulation of cyclin D1 during G₁-S phase progression and this effect was impaired in RASSF1A A133S cells (10). Thus, the A133S variant seems to be defective in cell cycle control.

Recently, Schagdarsurengin et al. showed a higher frequency of the A133S polymorphism in 141 Caucasian females with breast cancer compared with 70 individuals without breast cancer (15). However, the BRCA1/2 mutation status and the family history of breast cancer in these patients were not reported. In this study, our goal was to determine the contribution of RASSF1A A133S in breast cancer development in a larger patient population, with a particular focus on the subgroup of breast cancer patients with BRCA1/2 mutations.

Study population. Breast cancer patients and normal controls were all non-Ashkenazi Caucasian females. Informed consent was obtained from the controls and patients according to the University of Texas Southwestern Medical Center Institutional Review Board and national standards. Controls were women with no personal or family history of breast cancer. The majority of the controls were recruited in a population-based study (Genetic Epidemiology of Metabolic Syndrome, University of Texas Southwestern Medical Center). The rest of the controls were volunteers registered at the University of Texas Southwestern Medical Center. A self-administered questionnaire was used to collect information on individual's personal and family history of breast cancer. All of the control individuals fitting the criteria of no personal and family history of breast cancer were included in the study.

Breast cancer patients, including those seeking genetic counseling for breast cancer, were recruited from three institutes: 591 breast cancer patients were recruited sequentially from 1994 to 2006 at the Clinical Cancer Genetics Clinic at the University of Texas, Southwestern Medical Center. All patients from the University of Texas, Southwestern Medical Center (n = 591) signed consent forms allowing their DNA to be used for genetic research. Four hundred and seventy-one patients had their DNA analyzed for BRCA1/2 mutations and the probability for a BRCA1/2 mutation was estimated using the BRCAPRO model. The remaining 120 patients declined DNA analysis for BRCA1/2 mutations for various reasons including the financial cost or not wanting to know the result. Additional DNA from breast cancer patients with BRCA1/2 mutations were collected from the Dana-Farber Cancer Institute (n = 35) and from the University of Chicago School of Medicine (n = 27). Specimens collected at the Dana-Farber Cancer Institute were sequential women seen in the Dana-Farber Cancer Risk and Prevention and Breast Oncology Clinic, with germ line mutations in known breast cancer susceptibility genes (BRCA1/2, PTEN, TP53, LBK1, etc), and women at increased risk of breast cancer for other reasons (therapeutic thoracic radiation exposure, lobular carcinoma in situ, and other lesions). Specimens collected at the University of Chicago School of Medicine were from individuals who presented to the Cancer Risk Clinic at the University of Chicago between 1992 and 2007.

DNA extraction. Genomic DNA was extracted from blood by phenol/chloroform extraction and Proteinase K method or by using QIAamp DNA Mini Kit (QIAGEN).

Mutation detection. BRCA1/2 mutations were mostly identified by sequencing performed by Myriad Genetics. The entire coding region was sequenced in most of the patients except in a few cases in which mutation in a single site was determined based on a known BRCA1/2 mutation in other family members.

Genotyping. Fluorogenic 5′-nucleotidase assay in the TaqMan system (Applied Biosystems) was used to detect A133S polymorphism. Forward and reverse primers for amplifying RASSF1A are 5′GCCTGTGGAGTGGGAGACA and 5′GGGCATTGTACTCCTTGATCTTCT separately. Fam-labeled MGB probe 5′CCTTTCTCAAtCTGAG and Vic-labeled MGB probe 5′CCTTTCTCAAgCTGAG was used for detecting A133S and wild-type alleles separately. The assay was performed on a 7900HT Fast Real-Time PCR instrument. The reaction conditions were: 95°C for 10 min, followed by 40 cycles of 94°C for 15 s, and 60°C for 90 s.

Statistical analysis. Analysis was performed using SAS 9.1.3 Service Pack 3. A two-sample t test was used for comparing age differences and a χ² test was used for comparing allele frequencies among groups. In the cases in which the χ² test could not be performed, Fisher's exact test was used. Kaplan-Meir survival curves were constructed and a log-rank test was used to compare age at onset of breast cancer between A133S patients and wild-type RASSF1A patients in the BRCA1/2-positive and BRCA1/2-negative groups, respectively. P ≤ 0.05 was considered statistically significant.

Frequency of RASSF1A A133S in patients with breast cancer. We designed a high-throughput fluorogenic 5′-nucleotidase assay to determine the frequency of the A133S polymorphism in the RASSF1A gene (Supplemental data Fig. S1). Six hundred and fifty-three breast cancer probands and 190 controls were analyzed for the presence of A133S (Table 1). As shown in Table 2, 149 (22.8%) of the 653 breast cancer patients were homozygous or heterozygous for A133S. In contrast, 28 (14.7%) of 190 controls were homozygous or heterozygous for A133S [P = 0.017; odds ratios (OR), 1.71; 95% confidence intervals (95% CI), 1.10–2.66]. Thus, there is a significant difference between the frequencies of A133S in breast cancer patients that we studied compared with the controls.

In order to define the breast cancer risk associated with A133S, we divided the breast cancer patients into three groups: breast cancer patients with (n = 138) or without BRCA1/2 mutation (n = 395), and patients not tested for BRCA1/2 mutation (n = 120). As shown in Table 2, there is a higher frequency of A133S in BRCA1/2 mutation carriers compared with the normal controls (23.9% versus 14.7%, P = 0.037; OR, 1.82; 95% CI, 1.04–3.18).

To determine whether the A133S polymorphism is associated with familial breast cancer independent of BRCA1/2 status, we further divided the non-BRCA1/2 mutation carriers into two groups [higher familial breast cancer risk group (n = 223) and lower familial breast cancer risk group (n = 172)] using BRCAPRO, a computer model designed to calculate the predisposition of an autosomal dominant familial breast cancer phenotype (16, 17). As shown in Table 2, A133S is more frequent in patients in the higher familial breast cancer risk group (23.3%, P = 0.029; OR, 1.76; 95% CI, 1.06–2.92) compared with the controls. These results suggest that the presence of A133S is associated with familial breast cancer in general, and is independent of BRCA1/2 mutation status.

The presence of RASSF1A A133S is associated with a younger age at diagnosis in BRCA1/2 mutation carriers. BRCA1/2 mutation carriers develop breast cancer at an earlier age than average breast cancer patients without BRCA1/2 mutations. In our data, the median age of onset of breast cancer for all breast cancer cases was 43.0 years, which was 41.0 for BRCA1/2 mutation carriers. It is noted that the mean age at onset of breast cancer in our study group is significantly younger than the average for all breast cancer patients because of a study population bias toward patients with a family history of breast cancer. However, when we compared the age at onset among the BRCA1/2 mutation carriers with or without A133S, we found that A133S carriers were diagnosed 6 years earlier than those with wild-type RASSF1A (36.0 versus 42.0 years old, P = 0.002; Table 3; Fig. 1). In contrast, when we compared the age at onset among non-BRCA1/2 mutation carriers with or without A133S, we found that there was no significant age difference (42.0 versus 43.0 years old, P = 0.701; Table 3; Fig. 1). Thus, the presence of an A133S polymorphism along with a BRCA1/2 mutation in the same individual is associated with earlier onset of breast cancer.

In this study, we examined whether there was an association between the frequency of the RASSF1A A133S polymorphism and the incidence of sporadic and familial breast cancers. We show that there is higher frequency of RASSF1A A133S in breast cancer patients compared with the controls, which is consistent with prior work (15). In addition, we show that there is a higher frequency of RASSF1A A133S in familial breast cancers and in breast cancer patients carrying BRCA1/2 mutations. Our most significant finding is that patients with both A133S and BRCA1/2 mutations were diagnosed with breast cancers 6 years earlier than patients with the BRCA1/2 mutation only.

LOH and promoter hypermethylation are the two main mechanisms for RASSF1A inactivation in breast cancer. The results of this study suggest that the presence of a polymorphic allele of RASSF1A is another mechanism involving RASSF1A in breast cancer development. A simple hypothesis is that women carrying the A133S allele already have one inherited inactivation event (haploinsufficiency) for RASSF1A. This hypothesis is consistent with the previously published data showing loss of RASSF1A function with the A133S polymorphism reported by ourselves and others (10, 18). The phenotype of Rassf1a knockout mice also supports this hypothesis in which both Rassf1a^−/− and Rassf1a^+/− mice showed increased tumor multiplicity and tumor size relative to control animals in response to carcinogen treatment (19). Thus, comprehensive studies including LOH, promoter methylation, and now, A133S polymorphism are needed to further understand how loss of RASSF1A contributes to cancer development.

Reports on BRCA1/2 modifier genes are scarce and controversial (6). To our knowledge, there are no genes that have been confirmed to modify BRCA1/2 mutation risk with respect to early onset of breast cancer. We report here that there is an increase of RASSF1A A133S in patients with BRCA1/2 mutations. Moreover, the average age at breast cancer diagnosis for BRCA1/2 mutation carriers with A133S is 6 years earlier than in patients with wild-type RASSF1A. Our study supports the hypothesis that RASSF1A modifies the risk of breast cancer in BRCA1/2 mutation carriers.

In conclusion, our results warrant a large-scale study to validate the effect of A133S in the development of breast and other types of cancers. If confirmed, RASSF1A A133S could be developed as one of the molecular biomarkers for refining cancer risk estimates for women at higher risk for breast cancer. For individuals with a BRCA1/2 mutation, the confirmation of the association of A133S with earlier onset of breast cancer may indicate that breast cancer surveillance should start at an earlier age if they carry the A133S allele in RASSF1A.

Grant support: The Susan G. Komen Breast Cancer Foundation grant BCTR0402515 (B. Gao), Specialized Program of Research Excellence grants CA71618 and P50 CA70907, and the Margot Rosenberg Pulitzer Foundation (J.D. Minna).

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.

We thank Kristin Shelby for her excellent data management work.