Recurrent BRCA1 and BRCA2 Mutations in Mexican Women with Breast Cancer

This article is featured in Highlights of This Issue, p. 481

In Mexico, breast cancer has overtaken cervical cancer as the leading cause of cancer-related death in women (1–4) and mortality rates are increasing (5–7). Typically, breast cancer is diagnosed at a relatively advanced stage (III and above; refs. 8, 9) when the chance of cure is reduced. The median age of breast cancer diagnosis is 51 years (approximately one decade younger than that of women in Europe or North America) and almost one-half of Mexican women are premenopausal at breast cancer diagnosis (6–8).

The BRCA1 (10) and BRCA2 (11) genes account for between 5% and 10% of all breast cancer cases, and particularly in those women with a family history of breast and ovarian cancer (12–15), but the prevalence of mutations in these genes in Latin American women and their contribution to breast cancer is largely unknown. The lifetime risk of breast cancer in women who carry a BRCA1 or BRCA2 mutation is about 80% (12–14) but the absolute risk varies by country and by ethnic group (16). Characteristics of hereditary breast cancer include a young at age at onset and multiple cases of early-onset breast cancer or ovarian cancer in the family (12–14). However, as many as 50% of patients with breast cancer with an inherited mutation in BRCA1 and BRCA2 do not have a close relative with breast or ovarian cancer, either because their mutation is paternally inherited, their family is small, random segregation, and incomplete penetrance (17).

The prevalence of BRCA1 and BRCA2 mutations combined is approximately 0.3% in North America (12–17), but may be higher than this in countries or populations where there are founder mutations, such as Israeli Jews (18–20), Dutch (21), French-Canadians (22), Icelandic (23), Greenlandic (24), Polish (25), Russia and Eastern European (26–30), and Greek populations (31, 32). Recurrent mutations have also been described in women of Hispanic origin in the United States (33–38). The presence of recurrent BRCA1 and BRCA2 mutations has been noted in a few studies from Latin America and the Caribbean (39–42) including four small studies from Mexico (43–46). The identification of recurrent mutations greatly facilitates genetic testing of BRCA1 and BRCA2 (47). In this study, we screened for 26 mutations that have been observed previously in Mexican women and we have screened for other mutations in exon 11 in BRCA1 and exons 10 and 11 in BRCA2 in 810 Mexican women with breast cancer.

A multicenter breast cancer case–control study was established in 12 hospitals in three cities in Mexico (Mexico City, Monterrey, and Veracruz). DNA and epidemiologic data have been collected from January 2007 through June 2010 (48–51). Table 1 provides some of the descriptive statistics of 810 cases. A full summary of features of the cases and controls has been published elsewhere (48–51). The controls are not the focus of this current study. This study was designed to examine predictors of breast cancer risk among women ages 35 to 69 years. Cases were histologically confirmed new diagnosis of breast cancer, including invasive and in situ tumors. Data collection included the administration of a structured questionnaire by means of a face-to-face interview and anthropometric measurements and collection of a blood sample at the hospital by a trained nurse. All participants provided a written informed consent. The study was approved by the Institutional Review Board at each participating institution. The health questionnaire collected information on sociodemographic characteristics, reproductive factors, use of oral contraceptives and hormone replacement therapy, family and personal history of chronic diseases, personal history of transmitted sexual diseases, histories of body size, smoking, and alcohol consumption, and history of medical X-rays and mammograms. Subjects were informed of the goals of the study and the implications of the possible identification of a mutation in either BRCA1 or BRCA2. Subjects were permitted to decline participation for genetic testing.

Once a subject agreed to participate in the study, the research nurse collected blood samples in two 5 mL EDTA tubes. Blood samples were stored at each hospital at −20°C to −70°C and within 3 weeks, they were sent to the Instituto Nacional de Salud Pública, Cuernavaca, Mexico and stored at −70°C until shipment. The frozen blood was shipped on dry ice to the Narod laboratory in Toronto. Genomic DNAs were extracted from blood using the ArchivePure DNA Blood Kit (5Prime) according to the protocol. Stock DNA samples were bar-coded with a unique subject identification number to ensure reliable sample processing.

All samples were screened for 26 mutations found in the Mexican population; 21 in BRCA1 (MIM113705) and five in BRCA2 (MIM600185). Exon 11 of BRCA1 and exons 10 and 11 of BRCA2 were screened by the protein truncation test, PTT (TNT T7 Coupled Reticulocyte Lysate System, Promega; and [³⁵S] Methionine/Cysteine, New England Nuclear). Overlapping primer sequences were obtained from the Breast Cancer Information Core (BIC). PTT screening covered the three exons encompassing 17 known Mexican mutations in BRCA1 (K654X, 943ins10, S955X, Q1200X, R1203X, 1205del56, c.3124_3133delAGCAATATTA, c.2805_2808delAGAT, C1787S & G1788D, 2415delAG, 2525del4, 2552delC, 2925del4, 5382insC, 3148delCT, 3787delTA, and 4184del4) and five known Mexican mutations in BRCA2 (Q742X, W2586X, c.5114_5117delTAAA, c.2639_2640delTG, and 3492insT), as well as other Hispanic mutations and any other novel deleterious mutations in these exons.

The four remaining BRCA1 mutations were tested by differing methods. A tetra-primer Amplification Refractory Mutation System assay was designed for the exon 13 R1443X mutation, a restriction fragment length polymorphism was designed for the exon 18 A1708E mutation, and a TaqMan Copy Number Variation (CNV) assay (Applied Biosystems Inc., Assay ID: Hs05509065_cn) was used to detect the BRCA1 ex9–12del large rearrangement. The binding site of the probe for TaqMan CNV assay was on exon 10 of BRCA1 gene. To confirm the mutations identified by the TaqMan CNV assay and also determining the extent of the deleted region, a Multiple Ligation-dependent Probe Amplification (MLPA) assay (MRC Holland Inc., Assay ID: P002) on 3500XL genetic Analyzer (Applied Biosystems Inc.) was used. The 185delAG mutation, commonly seen in the Hispanic women in the United States of apparent Mexican and Jewish ancestry, was coupled in a previously designed multiplex assay (52). In addition, we tested for the BRCA1 5382insC and BRCA2 6174delT mutations commonly seen in Jews and others of eastern European ancestry using the same rapid multiplex method (52). Mutation-positive controls were included in the assay. All primer designs and PCR conditions are available upon request. All deleterious mutations detected by all methods were confirmed by direct sequencing [BigDye Terminator v.3.1 Cycle Sequencing Kit; 3130xL Genetic Analyzer (Applied BioSystems)] according to the manufacturer's protocol.

To compare the characteristics by menopausal or mutation status, Kruskal–Wallis for continuous variables and χ² and Fisher exact tests for categorical variables were used. The differences were considered statistically significant when P < 0.05. All analyses were conducted using Stata v. 12.

Eight hundred and ten women with breast cancer were included in the study. Of the 810 women, 66% were from Mexico City, 22% from Monterrey, and 12% from Veracruz. The median age of diagnosis was 51.5 years and 334 (41% of total) were premenopausal. A family history of any cancer was reported in 34% of the postmenopausal women and in 23% of the premenopausal women (P = 0.001). However, the frequency of first-degree relatives with breast cancer was similar in both groups (Table 1).

Thirty-five mutations were identified in 34 of the 810 women (4.3%; Table 2). In BRCA1, four recurrent mutations and four private mutations were detected in 20 women. The exon 9–12 mutation was detected in 8 women. The exon 18 C5242A mutation was detected in 4 women and two mutations were seen twice (exon 11 2552delC and exon 13 C4446T). In BRCA2, two recurrent mutations and nine private mutations were detected in 14 women. Two BRCA2 mutations were seen three times each (exon 10 2024del5 and exon 11 C4339T). One woman harbored two BRCA2 mutations (exon 10 2024del5 and 4321insAA). No Jewish founder mutations (BRCA1 exon 2 185delAG and exon 11 5382insC and BRCA2 exon 11 6174delT) were detected. Eighteen of the women (53%) with a BRCA mutation were from Mexico City; 8 mutation carriers (23%) were from Monterrey, and 8 mutation carriers (23%) with nine mutations were from Veracruz (including the woman with two BRCA2 mutations). The BRCA2 exon 11 2024del5 mutation was only found in 3 women from Veracruz (Table 2).

The mean age of breast cancer onset was 43 years in BRCA1 carriers, 50.9 years in BRCA2 carriers, and 52 years in noncarriers (P < 0.001; Table 3). The prevalence of mutations was 11.8% for women diagnosed ages 30 to 39 years, 4.8% for women diagnosed ages 40 to 49 years, 3.4% for women diagnosed ages 50 to 59 years, and 1.6% for women diagnosed at 60 years or older.

A history of any cancer in a first-degree relative was reported in 55% of BRCA1 carriers, 46.7% of BRCA2 carriers, and 28.5% of noncarriers. However, breast cancer in a first-degree relative was seen in only 3 of 20 (15%) women with BRCA1 mutations and 2 of 15 (13.3%) women with a BRCA2 mutation and 53 of 775 (6.8%) of noncarriers. The prevalence of mutations by cancer family history is provided in Table 3.

European ancestry was not associated with either BRCA1 or BRCA2 carrier status (Table 3; ref. 51).

We conducted a breast cancer case–control collection in 12 hospitals in three cities in Mexico. Eight hundred and ten blood samples from women with breast cancer were collected of whom 334 (41%) were premenopausal and 476 (59%) were postmenopausal. Thirty-five mutations were identified in 34 of the 810 (4.3%) women tested including eight unique BRCA1 mutations in 20 women and 11 unique BRCA2 mutations in 14 women (Table 2).

Genetic testing for mutations in BRCA1 and BRCA2 has potentially important public health implications for the detection of high-risk individuals for whom targeted prevention and tailored management strategies can be implemented (53). The ability to offer genetic testing in Mexico on a widespread level would be enhanced with the identification of common mutations in the two genes so the cost of genetic sequencing is reduced. In the present study, we detected recurrent mutations in 2.7% of 810 unselected cases of breast cancer. Twenty-two of the 34 mutation carriers had a mutation that was seen more than once, therefore the strategy of looking solely for recurrent mutations would have a sensitivity of approximately 60%. Ideally, to maximize sensitivity, one would screen all patients with breast cancer for both BRCA1 and BRCA2 in their entirety. However, given the current high costs of sequencing, this strategy is prohibitively expensive in Mexico. Alternate strategies include the testing of all high-risk patients for all mutations through full gene sequencing or testing all patients with cancer (high and low risk) for a smaller number of mutations (recurrent and founder mutations). Of interest, in the present study, only 3 of 20 (15%) women with BRCA1 mutations (all with the exon 9–12 deletion) and 2 of 14 (14.3%) women with a BRCA2 mutation (one harboring a 3036delACAA and another with 5770delA mutation) had a first-degree relative with breast cancer (Table 3) and, therefore, the strategy of testing only familial cases of breast cancer would result in the identification of only a minority of mutation carriers, even if complete sequencing were done for both genes. The 3 women with the BRCA1 exon 9–12 deletion developed breast cancer at 44, 44, and 56 years, respectively. The 2 women with BRCA2 mutations were <50 years at diagnosis (one was 42 years and the other woman was 48 years).

There are small reports of BRCA1/2 mutation screening studies in Mexico. Ruiz-Flores and colleagues (43) identified one BRCA1 3857delT and one BRCA2 2663-2664insA mutation among 51 Mexican patients with breast cancer (6% of 32 early-onset breast cancer patients). Vidal-Millán and colleagues (44) found three mutations in BRCA1 and BRCA2 genes in 40 Mexican patients with breast cancer (5%). Calderón-Garcidueñas and colleagues (45) found one BRCA1 mutation (exon 11, 3587delT) and one BRCA2 mutation (exon 11, 2664insA) in 22 early-onset Mexican breast cancer patients. Vaca-Paniagua and colleagues (46) found four mutations in 39 Mexican breast-ovarian cancer families, three of which were novel including BRCA1 c.3124_3133delAGCAATATTA and c.2805_2808delAGAT and BRCA2 c.5114_5117delTAAA and c.2639_2640delTG. Using our mutation screening strategy including PTT for exon 11 in BRCA1 and exons 10 and 11 in BRCA2, we would have detected each of the above mutations, were they present among the 810 Mexican women. However, we did not detect any of the above mutations.

Of note, no Jewish founder mutations were reported in any of these four studies, nor did we detect women harboring these mutations in this study. However, mutation screening studies performed in Latina women, mainly of Mexican origin in the United States revealed the presence of the Jewish BRCA1 exon 2, 185delAG founder mutation. Vogel and colleagues reported that 4 of 78 Hispanic women with familial breast cancer carried this mutation and 10 carried other mutations including BRCA1 2552delC (37). John and colleagues found a BRCA1 mutation in 21 of 393 (5.3%) of Hispanic women with breast cancer in California (38). They found the prevalence of BRCA1 mutation carriers of 3.5% [95% confidence intervals (CI), 2.1%–5.8%) in Hispanic patients (n = 393), compared with 8.3% (95% CI, 3.1%–20.1%) in Ashkenazi Jewish patients (n = 41) and 2.2% (95% CI, 0.7%–6.9%) in other non-Hispanic white patients (n = 508). The BRCA1 185delAG was the most common mutation in Hispanics and was found in five of 21 carriers (24%). Weitzel and colleagues (33) studied 110 unrelated Latina women at high risk of breast/ovarian cancer in Los Angeles. Thirty-four of the 110 women had a mutation (31%); of these 18 were of Mexican descent. Four mutations were seen more than once in women with Mexican origins: BRCA1 exon 2 185delAG (four times), exon 13 C4446T (R1443X; three times), exon 11 2552delC (two times), and BRCA2 exon 11 3492insT (two times). In a more recent follow-up study (34), the BRCA1 exon 13 C4446T (R1443X) was reported six times, four of which were in families of Mexican descent. The 2552delC was reported in four families and the A1708E was observed in three families of Mexican origin. The latter was also reported by Myriad Genetics in seven Latin American subjects in the BIC database (54). The BRCA2 exon 11 3492insT was identified in 10 families of Mexican descent only. In this study, we also identified the BRCA1 exon 13 C4446T (R1443X) twice; the exon 11 2552delC twice, the A1708E four times, and BRCA2 exon 11 3492insT once. Weitzel and colleagues (33) also reported single BRCA1 mutations which we also detected in the current study, each in a single individual: 1135insA, 2415delAG, and C3717T (Q1200X). We found recurrent BRCA2 mutations; exon 10 2024del5 (which was reported 11 times in BIC; ref. 54, but not in individuals of Latin American descent) and exon 11 C4339T (reported once in BIC; ref. 54, in an individual of Spanish descent) in three cases each. We identified a BRCA2 3036del4 mutation in a single case which was also identified by Osorio and colleagues (55) in a Spanish breast cancer family, while the 2024del5 mutations appears to be of Greek origin (31, 32). In summary, the BRCA1 1135insA, 2415delAG, 2552delC, C3717T, C4446T, C5242A and 9–12del and BRCA2 2024del5, 3492insT, and C4339T mutations are recurrent mutations in the Mexican population. One novel BRCA1 mutation (2190delA) and five novel BRCA2 mutations (2971del5, 4321insAA, 4534delAT, 5859delC, and 6686delC) were also identified in our study, which were not previously reported.

However, we did not find any BRCA1 exon 2 185delAG mutations in agreement with the four previous reports in Mexican patients with breast cancer (43–46) but in contrast with the observations in Hispanic American patients of Mexican descent (33–37). We included Jewish mutation-positive controls in each multiplex assay, in which the appropriate mutations were detected so we do not consider this result to represent a false negative. Possible explanations for the high frequency of the BRCA1 185delAG mutations in previous reports may be Spanish admixture of the population of Hispanic Americans in the United States from which the study subjects are drawn (56). Velez and colleagues investigated the ancestral origin of 33 unrelated individuals of Spanish descent with BRCA1 c.185delAG in Colorado (57). The presumed European component showed enrichment for Sephardic Jewish ancestry, consistent with historical accounts of Jewish migration from the realms that comprise modern Spain and Portugal during the Age of Discovery (58).

Weitzel and colleagues (34) also reported a founder deletion (BRCA1 exons 9–12) in 3.8% unrelated breast cancer families of Mexican origin. A recent follow-up study has shown that the BRCA1 ex9–12del deletion represents 10% of all BRCA1 mutations in 746 Hispanics with a personal or family history of breast and/or ovarian cancer and 492 population-based Hispanic breast cancer cases (35). We detected this mutation in 8 women (1% of the 810 women tested) or 22% of all observed BRCA1/2 mutations.

Overall, the age of breast cancer onset was around 8 years younger in BRCA1 carriers (43 years) compared with BRCA2 carriers (50.9 years) and 9 years younger than noncarriers (52 years; P < 0.001; Table 3).

One woman with breast cancer at age 53 years harbored two BRCA2 mutations (exon 10 2024del5 and exon 11 4321insAA). The exon 10 2024del5 mutation is a common mutation in Greek breast cancer families (31) and was found only in 3 women from Veracruz in this study. Biallelic BRCA2 mutations have been reported in Fanconi anemia (FA), specifically subtype D1 (59). FA is a recessive condition associated with progressive bone marrow aplasia, congenital abnormalities, and predisposition to leukemia and solid tumors of the head and neck, esophagus, and vulva (60). The biallelic BRCA2 mutations form of FA is severe with high risks of childhood cancer, particularly Wilms' tumor, brain tumors, and acute myelogenous leukemia (59–61). Recently, a woman with ovarian cancer was found to harbor biallelic BRCA1 mutations (62). To our knowledge, this is the first report of breast cancer in a biallelic BRCA2 mutation carrier. The combinations of BRCA2 mutations that are viable are limited. The exon 11 4321insAA has not been reported previously in BIC (54) or the literature and the functional impact of this mutation is unknown. Of note the Mexican woman with biallelic BRCA2 mutations in this study did not show signs of FA so the 4321insAA mutation may be nondeleterious.

Medullary breast cancer has been highly associated with BRCA1 mutations (63). In this study, we found that two of seven (28.5%) of the medullary breast cancers were from women with BRCA1 mutations. Although medullary breast cancer histologic subtypes represent a small number of the total number of breast cancers, the presence of this subtype is potentially an indicator for the presence of a BRCA1 mutation in Mexican women.

There are several strengths to this study. This is the largest study of BRCA1/2 mutations in Mexican women with breast cancer. The study population is not defined under the broad generational classification of “Hispanic,” rather a narrower definition of Mexican ancestry only. The lack of the Jewish founder mutations in this, and other published studies on Mexican breast cancer cases show the importance of studies in women in Latin American countries and that more attention should be paid to more clearly define the ancestral origin of “Hispanic” women in the United States.

This study also has limitations. The samples were not fully screened for mutations in the BRCA1 and BRCA2 genes via sequencing or dosage analysis such as MLPA. It is entirely plausible that with full screening of the genes, additional recurrent and common mutations may have been detected. Testing of a panel of recurrent mutations would be pragmatic but larger studies would be required to more definitively delineate a better set of true Mexican founder mutations. Furthermore, there is the possibility that some of the women in the study are related which may inflate the frequency of particular mutations.

Another limitation is the inability to consider second-degree family relatives and that the presence of ovarian cancer was not adequately recorded. However, family history in second-degree relatives is considered less predictive of BRCA-carrier status. Indeed, our results suggest that family history of breast cancer in first-degree relatives is not particularly predictive of BRCA-carrier status in Mexican women either.

In conclusion, studies of this kind are essential to determine the genetic etiology of breast cancer in Mexican women. These results highlight the variability in the mutation spectrum, penetrance, and phenotype of BRCA1 and BRCA2 mutations in Mexican women and reveal the presence of particular recurrent mutations in this population. Further comprehensive evaluation of the prevalence of BRCA1 and BRCA2 mutations is necessary. Through judicious testing of women believed to be at high risk for early-onset breast cancer, it is possible to identify highly predisposed women before the development of cancer. Current preventive options such as preventive mastectomy or tamoxifen may be tailored to the BRCA1/2 mutation carrier so as to improve morbidity and mortality associated with this disease.

No potential conflicts of interest were disclosed.

Conception and design: G. Torres-Mejía, M.R. Akbari, A. Angeles-Llerenas, C.M. Phelan, S.A. Narod

Development of methodology: G. Torres-Mejía, A. Angeles-Llerenas, E. Lazcano-Ponce, C.M. Phelan, S.A. Narod

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): G. Torres-Mejía, A. Angeles-Llerenas, C. Ortega-Olvera, E. Ziv, S.A. Narod

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): G. Torres-Mejía, R. Royer, M.R. Akbari, L. Martínez-Matsushita, C. Ortega-Olvera, E. Lazcano-Ponce, C.M. Phelan, S.A. Narod

Writing, review, and/or revision of the manuscript: G. Torres-Mejía, R. Royer, M.R. Akbari, A.R. Giuliano, C. Ortega-Olvera, E. Ziv, E. Lazcano-Ponce, C.M. Phelan, S.A. Narod

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): G. Torres-Mejía, R. Royer, M. Llacuachaqui, A. Angeles-Llerenas, S.A. Narod

Study supervision: G. Torres-Mejía, M.R. Akbari, A. Angeles-Llerenas, S.A. Narod.

The authors thank all the study participants, CONACyT for the financial support provided for this study, and all physicians responsible for the project in the different participating hospitals: Dr. Germán Castelazo (IMSS, Hospital de la Raza, Ciudad de México, DF), Dr. Sinhué Barroso Bravo (IMSS, Hospital Siglo XXI, Ciudad de México, DF), Dr. Fernando Mainero Ratchelous (IMSS, Hospital de Gineco-Obstetricia No 4. “Luis Castelazo Ayala,” Ciudad de México, DF), Dr. Hernando Miranda Hernández (SS, Hospital General de México, Ciudad de México, DF), Dr. Joaquín Zarco Méndez (ISSSTE, Hospital 20 de Noviembre, Ciudad de México, DF), Dr. Edelmiro Pérez Rodríguez (Hospital Universitario, Monterrey, Nuevo León), Dr. Jesús Pablo Esparza Cano (IMSS, Hospital No. 23 de Ginecología, Monterrey, Nuevo León), Dr. Heriberto Fabela (IMSS, Hospital No. 23 de Ginecología, Monterrey, Nuevo León), Dr. José Pulido Rodríguez (SS, Hospital Metropolitano Dr. “Bernardo Sepulveda,” Monterrey, Nuevo León), Dr. Manuel de Jesús García Solís (SS, Hospital Metropolitano Dr “Bernardo Sepúlveda,” Monterrey, Nuevo León), Dr. Fausto Hernández Morales (ISSSTE, Hospital General, Veracruz, Veracruz), Dr. Pedro Coronel Brizio (SS, Centro Estatal de Cancerología “Dr. Miguel Dorantes Mesa,” Xalapa, Veracruz), Dr. Vicente A. Saldaña Quiroz (IMSS, Hospital Gineco-Pediatría No 71, Veracruz, Veracruz), and Dr. PH. Teresa Shamah Levy (INSP, Cuernavaca Morelos) and Ma. del Pilar Cuellar-Rodríguez (INSP, Cuernavaca Morelos).

E. Ziv was recipient of the NIH grant R01CA120120 and K24CA169004.

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.