Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Comprehensive Analysis of BRCA1, BRCA2 and TP53 Germline Mutation and Tumor Characterization: A Portrait of Early-Onset Breast Cancer in Brazil

Abstract

Germline mutations in BRCA1, BRCA2 and TP53 genes have been identified as one of the most important disease-causing issues in young breast cancer patients worldwide. The specific defective biological processes that trigger germline mutation-associated and -negative tumors remain unclear. To delineate an initial portrait of Brazilian early-onset breast cancer, we performed an investigation combining both germline and tumor analysis. Germline screening of the BRCA1, BRCA2, CHEK2 (c.1100delC) and TP53 genes was performed in 54 unrelated patients <35 y; their tumors were investigated with respect to transcriptional and genomic profiles as well as hormonal receptors and HER2 expression/amplification. Germline mutations were detected in 12 out of 54 patients (22%) [7 in BRCA1 (13%), 4 in BRCA2 (7%) and one in TP53 (2%) gene]. A cancer familial history was present in 31.4% of the unrelated patients, from them 43.7% were carriers for germline mutation (37.5% in BRCA1 and in 6.2% in the BRCA2 genes). Fifty percent of the unrelated patients with hormone receptor-negative tumors carried BRCA1 mutations, percentage increasing to 83% in cases with familial history of cancer. Over-representation of DNA damage-, cellular and cell cycle-related processes was detected in the up-regulated genes of BRCA1/2-associated tumors, whereas cell and embryo development-related processes were over-represented in the up-regulated genes of BRCA1/2-negative tumors, suggesting distinct mechanisms driving the tumorigenesis. An initial portrait of the early-onset breast cancer patients in Brazil was generated pointing out that hormone receptor-negative tumors and positive familial history are two major risk factors for detection of a BRCA1 germline mutation. Additionally, the data revealed molecular factors that potentially trigger the tumor development in young patients.

Citation: Carraro DM, Koike Folgueira MAA, Garcia Lisboa BC, Ribeiro Olivieri EH, Vitorino Krepischi AC, de Carvalho AF, et al. (2013) Comprehensive Analysis of BRCA1, BRCA2 and TP53 Germline Mutation and Tumor Characterization: A Portrait of Early-Onset Breast Cancer in Brazil. PLoS ONE 8(3): e57581. https://doi.org/10.1371/journal.pone.0057581

Editor: Paolo Peterlongo, IFOM, Fondazione Istituto FIRC di Oncologia Molecolare, Italy

Received: October 17, 2012; Accepted: January 22, 2013; Published: March 1, 2013

Copyright: © 2013 Carraro et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Funding: This work was supported by grants from the National Institute of Science and Technology in Oncogenomics (Fundação de Amparo à Pesquisa do Estado de São Paulo - FAPESP 2008/57887-9; Conselho Nacional de Desenvolvimento Científico e Tecnológico - CNPq 573589/08-9), Centro de Pesquisa, Inovação e Difusão - CEPID (FAPESP 98/14335-2), and FAPESP 2009/10088-7. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Breast cancer in patients under the age of 35 y occurs in 2–10% of cases in Western countries, although this frequency may differ among different ethnic groups [1]–[5].

In Brazil, the incidence of breast cancer is high, with a trend of increased incidence among younger women since the 1980s. In the age range of 25–29 y, the rate increased from 6.4 to 7.8 per 100,000 women, while in the range of 30–34 y, the rate of incidence increased from 19 to 27.6 per 100,000 women [6]. This boost in early-onset breast cancer may be explained by either an increase in case notification or as a result of changes in the exposure pattern to different environmental risk factors [6]. Early-onset breast cancer is associated with worse outcome, despite aggressive therapies [1], [4], [5], [7]–[9]. Accordingly, invasive breast carcinomas in young patients exhibit clinical-biological characteristics of aggressive disease [8]–[11] and are associated with poor relapse-free survival [12]. This phenomenon can be partially attributed to the greater frequency of hormonal receptor/HER2-negative tumors in this group compared with late-onset breast cancer patients [12] in addition to poor differentiation, lymphovascular invasion and high proliferative fraction [10], [13].

Breast cancer has increasingly been described as a heterogeneous disease that displays a variety of subtypes with distinct gene expression profiles that have substantial implications for prognoses and survival rates [14]. It has been suggested that biological differences in tumors of early- and late-onset breast cancer patients are mainly influenced by expression profiles inherent to breast cancer subtype and grade [15].

The risk factors for early-onset breast cancer patients are still poorly understood; however, a familial history of cancer is a very important feature present in 10–37% of all cases. Among early-onset familial cases, 10–40% was found to be associated with BRCA1 and BRCA2 (BRCA1/2) mutations. In contrast, among sporadic early-onset breast cancer patients, the frequency of BRCA1/2 mutation ranges from 1–10% [16]–[18]. Other susceptibility genes for breast cancer, such as TP53, ATM, PALB2, and the deletion at position 1100 of the CHEK2 gene account for a small proportion of familial breast cancer patients [19]. Compelling data have shown that breast tumors from patients carrying germline BRCA1/2 mutations are also morphologically and genetically different from each other as well as both sporadic and hereditary BRCAx-associated tumors. The last category is a heterogeneous group supposedly driven by mutation in as-yet unidentified genes [20]–[24].

The specific defective biological processes that trigger BRCA1/2-associated and -negative tumors remain unclear; whether tumorigenesis in early- and late-onset breast cancer patients differs is also unknown. Therefore, our main goals in the current study were to determine the mutation rate of the major breast cancer susceptibility genes in young Brazilian breast cancer patients and to characterize the immunohistochemical and molecular features of their tumors. We screened the BRCA1, BRCA2, CHEK2 (c.del1100C) and TP53 genes for germline mutations in a cohort of 54 young women under the age of 35 y who developed breast cancer. We investigated their respective tumors with respect to hormonal receptors and HER2 status and compared the results with a cohort of 224 tumors of late-onset breast patients. We also assessed the transcriptional profiles of the tumors of the early-onset breast cancer patients. Additionally, we investigated the pattern of germline copy number variations (CNVs) and somatic acquired chromosomal alterations (SCNA) in a subset of matched samples. Taken together, the results permitted the outlining of a portrait of early-onset breast cancer in Brazil.

Materials and Methods

Patients

Patients were ascertained at three reference cancer centers in the state of São Paulo, Brazil: Hospital A. C. Camargo, São Paulo; Instituto Brasileiro de Controle do Câncer, IBCC, São Paulo; and Hospital do Câncer de Barretos, Barretos. All patients provided a written informed consent agreeing in participating in this study. All patients received genetic counseling. This study was approved by the Institutional Ethics Committee under number 818/06 (AC Camargo Hospital).

Fifty-four unrelated young patients with breast cancer diagnosed at an early age (35 y) were included in the study for germline mutation screening. The patients were classified on family history based on NCCN (www.nccn.org) criteria for Breast and Ovarian Cancer Syndrome. Tumor and blood samples were collected during biopsy or breast surgery. Peripheral blood of an affected sister diagnosed with breast cancer an age of 29 was used for confirming the germline alteration identified in the index patient. Patients received no neoadjuvant treatment before tumor and blood collection, with the exception of patient ID_2019. Two samples of peripheral blood (5 ml) were collected; fresh frozen tumor samples were submitted to histological analysis and manual dissection was performed by a pathologist. Only samples containing at least 70% malignant cells were included in the study. An additional group of tumor samples was derived from a cohort of 224 female patients diagnosed at ≥50 y.

All formalin-fixed paraffin-embedded tissues were tested for estrogen (ER) and progesterone receptor (PR) and HER2 expression by immunohistochemistry in tumors derived from the 55 young patients (54 unrelated and one affected sister) and from the 224 women of the additional group (≥50y). HR positive was considered when either ER or PR was positive. FISH analysis was performed to detect HER2 amplification in tumor samples with a HER2 score 2+ detected by immunohistochemistry reaction. The HER2 status was classified as positive when HER2 (score 3+) was detected by immunohistochemistry or HER2 DNA amplification was detected by FISH analysis.

Methods

Full details of methods are given in the online Material and Methods S1.

Briefly, the coding regions including intron-exon boundaries of BRCA1 (U14680 or NM_007294.3), BRCA2 (U43746 or NM_000059.1) and TP53 (NM_000546) genes were sequenced in both the forward and reverse directions, and CHEK2 (NM_007194.3) was screened for the c.1100delC mutation. Chromatographic tracings were analyzed using the CLC Bio software. Nucleotide alterations were searched in the BIC Database (Breast Information Core; http://research.nhgri.nih.gov/bic, freeze October, 2012). Genes were considered as wild type when the nucleotide missense alterations were classified as no clinical relevance in BIC database and/or as no or little clinical significance (values 1 and 2, respectively) in LOVD-IARC database. Genes were considered as unclassified variant (UV) when the nucleotide missense alterations were categorized as unknown clinical relevance in BIC and/or as uncertain in LOVD-IARC (value 3) database. In cases of disagreement between the two databases, the classification of LOVD-IARC was taken into consideration. Genes with any type of insertion or deletion or amino acid substitution that result in premature_stop codons before amino acids 1853 and 3309 within the BRCA1 and BRCA2, respectively, were classified as mutated. The UVs were submitted to in silico prediction programs. Nucleotide ambiguities leading to amino acid changes in the p53 protein were searched in the IARC database (International Agency for Research on Cancer; http://www-p53.iarc.fr/index.html). All detected alterations were confirmed in a second DNA sample in both the forward and reverse directions.

For gene expression analysis, tumor samples of the 55 young patients (54 unrelated patients and one affected sister) were included. One-color labeled cRNAs were hybridized to the Agilent B4X44K G4112F whole human genome oligoarray (Agilent, Santa Clara, USA). Data were analyzed with a permuted t-test (MEV, TM4 software), and genes were considered differentially expressed when p≤0.01 and fold-change ≥|2| (correction by adjusted Bonferroni method). Hierarchical clustering of samples was verified by Pearson correlation distance and complete linkage methods. Over-representation of pathways and biological process in the differentially expressed genes was determined with FunNet software (Functional Analysis of Transcriptional Networks), using KEGG and Gene Ontology (GO) annotations (level 9). All microarray raw data have been deposited in the GEO public database (http://www.ncbi.nlm.nih.gov/geo), a MIAME compliant database, under accession number GSE37126 (http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?token=bzqzlaugqkeqsle&acc=GSE37126).

Comparative genomic hybridization based on microarrays (array-CGH) was performed for investigating DNA copy number alterations using a 180 K whole-genome platform (Oxford Gene Technology, Oxford, UK) as previously described [25]. Germline array-CGH data were also visually inspected for copy number imbalances within the BRCA1, BRCA2, and TP53 genes in resolution of a single probe. The full germline DNA copy number data for the patients without BRCA1/2 mutations have been previously reported [25].

Results

Patient and Tumor Characteristics

For germline mutation screening 54 patients were included in the study (see Table S1 for complete information), with a median age of 31 y (range 22–35 y). Of the 51 unrelated patients interviewed, 16 (31.4%) reported positive familial history [FH(+)].

The majority of all young patients (89%) was diagnosed with invasive ductal carcinoma either of intermediate or high histological grades and early-stage disease (clinical stages I/II, 58%). Most tumors (76.4%) were hormonal receptor-positive [HR(+)] [76.4% ER(+) and 60.0% PR(+)], 20% presented positive HER2 status [HER2(+)] (one patient had unknown HER2 status), and 20% were triple-negative (TN) (complete information in Table S1).

Analysis of the Hormone Receptor and HER2 Status of Breast Tumors from Early-onset (≤35 y) and Late-onset (≥50 y) Patients

At first, we compared the protein expression of routinely used immunohistochemistry markers [ER/PR for hormonal receptors (HR)] and HER2 status in tumors from the 55 young (≤35 y) (54 unrelated and 1 sister) and old patients (50 y). The latter group comprised 224 patients with a median age of 64 y (50–93 y), all patients presented invasive ductal carcinomas (IDC).

No differences in the frequency of HR(+), HR(−), HER2(+) or TN tumors were detected between early-onset and late-onset breast tumor patient groups (this analysis considered only 49 tumors diagnosed as invasive ductal carcinoma in the group of young patients). Significant differences in high grade and advanced clinical stage frequencies were observed. High-grade tumors were significantly detected in young patients (p = 0.021), while advanced clinical stage tumors occurred more frequently in older patients (p = 0.031) (Table 1).

thumbnail
Download:
Table 1. Distribution of clinical and histopathological features in young and older patients (considering only IDC histological type).

https://doi.org/10.1371/journal.pone.0057581.t001

Frequency of BRCA1, BRCA2, TP53 and CHEK2 (c.1100delC) Mutations in Brazilian Patients ≤35 y

Deleterious mutations in BRCA1 and BRCA2 genes were found in 11 of the 54 (20.5%) of the unrelated patients. Thirty-two patients were classified as BRCA1/2 wild type (59%) and 10 as UV carriers (18.5%). Mutations were detected in 7 (13%) patients for BRCA1 gene and in 4 (7.5%) patients for BRCA2 (Table 2).

thumbnail
Download:
Table 2. Deleterious mutations detected in the BRCA1, BRCA2 and TP53 genes.

https://doi.org/10.1371/journal.pone.0057581.t002

Three of these BRCA2 mutations and one of BRCA1 were reported for the first time [BRCA2: p.Q756X, p.C1654X and c.4968insGT that results in a premature stop codon at amino acid 1617; BRCA1: c.560+2T>A, a splice-site variant that leads in an aberrant transcript with a premature stop codon (data not shown)].

In the 10 UV-carrier patients, 8 distinct missense alterations were identified (p.T1915M detected in two unrelated patients and p.I2490T in four unrelated patients) (Table 3). Three UVs have not been previously described (p.S1655P and p.A1669V in BRCA1 and p.D381G in BRCA2).

thumbnail
Download:
Table 3. Unclassified Variants (UVs) identified in BRCA1 and BRCA2 genes.

https://doi.org/10.1371/journal.pone.0057581.t003

The 1100-deletion in the CHEK2 gene was not found in any of the samples studied. Finally, 43 patients negative for BRCA1/2 pathogenic mutations were also screened for TP53, and only one was found to be mutated. This pathogenic alteration (p.V143M) has already been reported in a tumor as a somatic mutation in the IARC database.

Relationship between Mutation Status and Tumor Subtype and Familial History of Hereditary Cancer

Positive significant associations were observed between BRCA1/2-mutated carriers with both HR(−) and triple-negative (TN) tumors. No significant association was found between BRCA1/2-mutated carriers and HER2 status of tumors (Table 4).

thumbnail
Download:
Table 4. Distribution of BRCA1/2 status according to immunohistochemical characteristics and familial history.

https://doi.org/10.1371/journal.pone.0057581.t004

Patients reporting FH(+) had a significant higher probability of harboring BRCA1/2 mutation. Of the 10 unrelated_patients carrying BRCA1/2 mutations (7 in BRCA1 and 3 in BRCA2) for whom family history was known, 7 reported a positive familial history (70%). Of the 16 unrelated patients with FH(+), 37.5% carried pathogenic germline mutations in BRCA1 gene, against only 6.2% in BRCA2, revealing that FH(+) is one of the major risk factor for BRCA1 mutations in Brazilian young patients.

By evaluating the frequency of tumor subtypes as a function of the mutation status of individual genes, our data revealed that 6 out of 7 (85.7%) BRCA1 mutation-carriers developed HR(−) tumors; among them, 5 were TN (71.4%). All 5 patients who were BRCA2 or TP53 mutation-carriers developed HR(+) tumors.

Of the young unrelated patients with HR(−) tumors, 50% (6/12) harbored a deleterious germline mutation in the BRCA1 gene; this frequency was similar (5/10) in patients with TN tumors. In contrast, in unrelated patients diagnosed with HR(+) tumors, only 9.5% (4/42) and 2.4% (1/42) harbored a deleterious germline mutation in BRCA2 and TP53, respectively.

Finally, analysis of both tumor subtype and FH(+) revealed that 83% (5/6) and 80% (4/5) of patients with FH(+) diagnosed with HR(−) or TN tumors carried a BRCA1 mutation, respectively. No association was observed among patients with FH(+) diagnosed with HR(+) tumor subtype and mutations in BRCA2 or TP53 genes; of 10 patients, only one was a BRCA2 carrier (10%). The TP53-mutation carrier was a 24 y patient who did not report a family history of cancer.

Germline and Tumor Genomic Imbalances in Early-onset Breast Cancer Patients

We also assessed germline and somatic genomic imbalances in 15 patients (blood and tumor matched samples), 7 of which carried BRCA1/2 germline mutations (3 in BRCA1 and 4 in BRCA2), and 6 and 2 of which harbored BRCA1/2 wild type and BRCA1/2 UVs, respectively. This analysis had two basic purposes: first, to search for germline intragenic deletions and/or duplications in BRCA1, BRCA2, TP53 and CHEK2 genes; and second, to compare the number of somatic copy number alterations (SCNA) in tumors driven or not driven by BRCA1/2 germline mutations.

No germline deletions or duplications were observed in the four genes by array-CGH analysis. Additionally, BRCA1/2-mutated tumors did not exhibit a higher degree of genomic instability relative to wild-type- and UV-associated tumors, at least as measured by the total number of SCNAs (Table S2). However, we observed that 3 out of 4 tumors associated with germline BRCA2 mutations exhibited deletion of the BRCA2 gene in mosaic (IDs 2048, 2031 and 2025).

Gene Expression Analysis of Breast Tumors: Identification of Differentially Expressed Genes in BRCA1/2-associated and −negative Tumors

To identify a differential gene signature associated to BRCA1/2 deleterious mutation gene expression analysis was performed in 49 samples for which the tumors were available (10 tumors from BRCA1/2 mutated carriers; 28 from wild-type BRCA1/2 patients; 10 from BRCA1/2 UV-carriers, and 01 from TP53-mutated carrier). For this analysis, gene expression profile of 28 tumors from wild-type BRCA1/2 patients was compared with the 10 tumors from BRCA1/2-mutated carriers.

This analysis revealed 34 differentially expressed genes: 18 up-regulated in BRCA1/2-mutated tumors and 16 up-regulated in BRCA1/2-negative tumors (Table 5). To provide functional insights, we annotated these 34 genes in the biological process category of the Gene Ontology (GO) and KEGG pathways. Thirty-one genes could be categorized in GO (Tables S3 and S4). The over-represented Biological Process (GO) categories for genes up-regulated in the mutated tumors included mainly DNA damage, cellular and cell cycle-related processes. In contrast, the up-regulated genes in negative tumors were preferentially included in cell and embryo development-related processes (Figure S1). Within the KEGG categories, up-regulated genes in BRCA1/2-associated tumors were enriched for cell cycle pathways, mismatch repair, glutathione metabolism, oocyte meiosis and progesterone-mediated oocyte maturation. None of the KEGG categories were enriched in the group of genes up-regulated in BRCA1/2-negative tumors.

thumbnail
Download:
Table 5. Differentially expressed genes between BRCA1/BRCA2-negative and -positive mutation-driven tumors.

https://doi.org/10.1371/journal.pone.0057581.t005

Hierarchical clustering based on this set of genes discriminated 100% of the BRCA1/2-associated tumors from 75% of the BRCA1/2-negative tumors (Figure 1).

thumbnail
Download:
Figure 1. Hierarchical clustering based on 34 differentially expressed genes in BRCA1/BRCA2-associated and -negative tumors.

Each row represents a gene, and each column represents a tumor sample. Red indicates strong expression; green indicates weak expression; and black indicates moderate expression. Red squares represent BRCA1 or BRCA2 pathogenic-associated tumors, and green squares represent tumors from BRCA1/2 WT (non mutated). The colored lines of the dendrogram represent the support for each clustering: black and gray lines indicate greater reliability; yellow and red lines indicate lesser reliability.

https://doi.org/10.1371/journal.pone.0057581.g001

Next, we performed hierarchical clustering including the additional 10 tumors from patients carrying UVs in the BRCA1/2 genes and the TP-53 associated tumor (Figure S2). Interestingly, the clustering based on gene expression of the 49 tumor samples grouped 93% of the BRCA1/2-negative tumors discriminating from 100% of BRCA1/2-associated tumors. In regarding to UV breast tumor samples, 3 (30%) and 7 (70%) out of 10 samples were clustered with BRCA1/2-associated and -negative tumors, respectively. Tumors from two affected sisters (IDs 2007 and 2012) whose germline UV identified in the index patient (ID 2007) was confirmed in the sister (ID 2012) (BRCA1- p.S1655P) were discriminated into two different cluster ramifications; these two tumors were of different subtypes: one was TN, high-grade and atypical medullar, while the other was invasive ductal carcinoma, ER(+), HER(−) and of histological grade 2. The TP53-associated tumor clustered with the BRCA1/2-associated tumor group.

Combined Analysis of Gene Expression and Chromosomal Imbalances of Breast Tumors

The set of 34 genes identified as up- or down-regulated in the group of BRCA1/2-associated tumors was interrogated for DNA gains and losses. We considered a concordant pattern when at least two tumors in each group exhibited: a) gains for up-regulated genes in BRCA1/2-associated tumors and/or loss in -negative tumors, b) losses for down-regulated genes in BRCA1/2-associated tumors and/or gains in -negative tumors. A total of 8 genes displayed a concordant pattern namely FMR1 and TMPO (up-regulated), and SRCIN1, TCAP, ZNF396, IFT140, FRY and TRAF3IP1 (down-regulated) (Table 5). A discordant opposite pattern was not observed. The remaining genes were either not affected by SCNA or were randomly affected by gains and losses irrespective to the BRCA1/2 mutational status.

Discussion

An increased risk of death has been observed in young women affected by breast cancer [9], implying that tumors in early-onset cancer patients could be a distinct entity of breast cancer. Tumor aggressiveness in young women has been reported worldwide based on increases in the rate of high-grade, fast-proliferation, HR(−), basal-like and HER2-enriched breast tumors [1]–[3], [9], [13], [15], [26], [27] and their poorer overall and disease-free survival rates [5]. In the current study, tumor aggressiveness was assessed by comparing groups of tumors from younger and older women. Our results revealed a higher percentage of high histological tumor grade in the early-onset breast cancer group compared with the late-onset group, similar to other studies [4]. We did not detect an increase of HR(−) or TN breast tumors in young women, finding in contrast to those previously reported in the Brazilian [28]–[30] and other populations [1], [12] and in agreement with some studies [3], [4].

Inactivating mutations in cancer susceptibility genes, such as BRCA1 and BRCA2, which are inherited in an autosomal dominant pattern, are the major genetic factor associated with a high risk of breast cancer at an early age. The percentage of BRCA1/2 germline mutations in early-onset breast cancer patients ranges widely, from 11 to 24% in different studies [16], [18], [31]–[33]. Here, we reported a 20.4% rate of BRCA1/2 deleterious mutation, a frequency comparable to those described in Caucasian, Korean, American [<36 y (16.7%)] [18], British [<31 y (16%)] [16], Canadian [<36 y (16%)] [32] and Cypriot patients [<40 y (23%)] [33] and distinct to French patients [<36 y (10.9%)] [34].

Our data pointed out that BRCA1 mutation screening is mandatory for young Brazilian patients diagnosed with HR(−) and/or TN breast tumors, specially when it occurs in combination with FH(+), supporting previous studies that have reported an increased probability of BRCA1 germline mutation in_young patients with FH(+) and TN tumors [35]–[37]. It is well known that Brazilian population harbor a complex genetic background, reinforcing that both features, negative hormonal receptor tumors [HR(−) and/or TN] and FH(+), are very solid risk factors for BRCA1 mutation in young women, irrespective of their genetic composition. Nevertheless, an extensive evaluation of the prevalence of the BRCA1 mutation in TN and HR(−) tumors similar to that performed in the British population [37] is needed for proper genetic counseling of individuals and families at higher risk of breast cancer in Brazil.

Eighteen and a half percent of our patients (10 out of 54) presented BRCA1/2 UVs, and most of these patients were diagnosed with HR(+) tumors. The reported frequencies of BRCA UVs vary in different ethnic populations, with higher rates in African-American (38%) than in Caucasian (10%) and Korean patients (12%) [18]. The intermediary UV frequency in the patients in our study (18.5%) may reflect the high genetic miscegenation of the Brazilian population.

Among the eight types of UVs found in this study, the variant BRCA2: c.7697T>C, p.I2490T was detected in four distinct young patients, one of them is a carrier of a novel nonsense BRCA2 mutation (c.5190T>A - p.C1654X). This fact suggests low likelihood of this variant to play a deleterious function and consequently to be a disease-causing mutation.

Another important genetic factor related to early-onset breast cancer is the occurrence of germline TP53 mutations, which are associated with Li-Fraumeni Syndrome or Li-Fraumeni-like syndromes. In this cohort, a germline TP53 mutation was detected in only one case, in line with others studies that found very low frequencies of TP53 mutations (1%) in early-onset patients [38], [39]. Although the TP53 pR337H mutation was reported as a founder effect mutation in the population of southern Brazil [39] and has been detected at high frequency in Brazilian families with high cancer predisposition [40], we did not detected this specific mutation. This result can be attributed to the relatively low penetrance of this mutation for breast cancer in women below the age of 30 y [39].

No germline copy number alterations affecting the BRCA1, BRCA2, TP53 or CHEK2 genes were identified. A whole-genome investigation in Brazilian early-onset and FH(+) breast cancer patients detected rare germline CNVs [25]; one of the reported patients and her affected sister carried a 540 kb 1p31.1 microdeletion encompassing only 3 genes (ST6GALNAC3, ST6GALNAC5, PIGK); both patients were included in the present study. The most relevant gene in the affected region is ST6GALNAC5, a sialyltransferase recently identified as related to the development of breast cancer metastasis [41], suggesting a possible role for this gene in the development of the early-onset breast cancer in these patients.

Gene expression signatures have also been used for distinguishing breast tumor subtypes [14], chemotherapy-resistant and -sensitive samples [42], and pre-invasive lesions with distinct malignant potential [43], demonstrating that it is a very efficient approach for categorizing heterogeneous tumors. In the current study, we identified a transcriptional signature associated with BRCA1/2 status that distinguished BRCA1/2-associated tumors from negative tumors and suggested distinct biological processes involved in driving transformation in these tumor groups of young patients. The intrinsic molecular subtypes determined by gene expression profile strongly influence patient prognosis [44] and surely other important tumor characteristics. Three genes (RRM2, UBE2T and EXO1) belonged to the list of 50 genes associated to molecular subtype (PAM50) [45] were detected in the gene expression signature associated to BRCA1/2 status. Therefore, if the gene expression modulation of these three genes is really influenced by BRCA1/2 mutations or by the molecular subtypes is hard to be estimated.

Interestingly, 3 of the 8 genes exhibiting a concordant pattern in the genomic and transcriptional analysis (FMR1, SRCIN1 and TCAP)_are annotated in those over-represented categories, reinforcing the involvement of defective cellular- and embryo development-related processes in triggering breast tumorigenesis in BRCA1/2-associated and -negative groups, respectively. FMR1, up-regulated in BRCA1/2-associated tumors, is located in chromosome Xq27.3. The protein encoded by FMR1 binds RNA and seems to be involved in the traffic of mRNAs from the nucleus to the cytoplasm. Remarkably, mutation in this gene has been associated with ovarian cancer risk [44]. Both SRCIN1 and TCAP genes, up-regulated in BRCA1/2-negative tumors, are located in 17q12. SRCIN1 protein, also known as p140CAP, regulates the oncogene SRC kinase interfering in balance from SRC active to inactive [46].

p140CAP arrests E-cadherin at the cell membrane and prevents EGFR and Erk1/2 signaling, decreasing proliferation of tumor cells [47]. The protein encoded by TCAP_is found in striated and cardiac muscle and mutation in this gene has been associated with limb-girdle muscular dystrophy type 2G [48]. Although this gene is mapped in a region commonly amplified in breast tumor, nothing is known about its role in the tumor context. All three genes are promising candidates that deserve further investigation of their role in breast cancer, especially in the context of BRCA1/2 status.

The experimental approach, combining germline and somatic analysis, has shed light on some of the genetic factors that trigger the development of breast cancer at an early age, which will aid in establishing additional criteria for genetic testing. Altogether, data delineated an initial portrait of Brazilian early-onset breast cancer patients, contributing to the establishment of public health standards for referring patients for genetic testing and leading to more personalized and effective management of breast cancer in Brazil.

Supporting Information

Figure S1.

GO biological process-enriched categories of the up- and down-regulated genes in BRCA1/2-associated tumors. The bar corresponds to the percentage of differentially expressed genes in relation to all annotated genes in the respective category.

https://doi.org/10.1371/journal.pone.0057581.s001

(TIF)

Figure S2.

Hierarchical clustering based on 34 differentially expressed genes in BRCA1/BRCA2-associated and -negative tumors. Each row represents a single gene, and each column represents a tumor sample. Red indicates strong expression; green indicates weak expression; and black indicates moderate expression. Red squares represent BRCA1 or BRCA2 pathogenic-associated tumors, and green and blue squares represent tumors from BRCA1/2 non-mutated and unclassified variant carriers, respectively. Purple square represents tumor from TP53 mutated carrier. The colored lines of the dendrogram represent the support for each clustering: black and gray lines indicate greater reliability; yellow and red lines indicate lesser reliability.

https://doi.org/10.1371/journal.pone.0057581.s002

(TIF)

Table S1.

Clinical data of the patients included in the study.

https://doi.org/10.1371/journal.pone.0057581.s003

(DOC)

Table S2.

Summary of the array-CGH results of fifteen tumor samples.

https://doi.org/10.1371/journal.pone.0057581.s004

(DOC)

Table S3.

Up-regulated genes in BRCA1/2-associated and -negative tumors in the enriched GO Biological Process categories.

https://doi.org/10.1371/journal.pone.0057581.s005

(DOC)

Table S4.

Up-regulated genes in BRCA1/2-associated tumors distributed in the enriched categories of the KEGG pathway.

https://doi.org/10.1371/journal.pone.0057581.s006

(DOC)

Material and Methods S1.

Full details of methods.

https://doi.org/10.1371/journal.pone.0057581.s007

(DOC)

Acknowledgments

We are indebted with the patients and their families, and thank the A.C. Camargo Biobank. We also thank Elisa Napolitano e Ferreira and Mabel Pimilla Fernandez for technical assistance in microarray analysis.

Author Contributions

Ascertained and selected patients: MSM RADM ECL SHGG MIWA. Conceived and designed the experiments: DMC MAAKF HB MIWA CAMF MMB. Performed the experiments: DMC BCGL EHRO ACVK AFC LDCM FAS. Analyzed the data: DMC BCGL EHRO ACVK AFC LDCM FAS RDP HB. Contributed reagents/materials/analysis tools: DMC CAMF MMB. Wrote the paper: DMC MAAKF EHRO ACVK MMB.

References

  1. 1. Gonzalez-Angulo AM, Broglio K, Kau SW, Eralp Y, Erlichman J, et al. (2005) Women age<or = 35 years with primary breast carcinoma: disease features at presentation. Cancer 103: 2466–2472.
  2. 2. Uhrhammer N, Abdelouahab A, Lafarge L, Feillel V, Ben Dib A, et al. (2008) BRCA1 mutations in Algerian breast cancer patients: high frequency in young, sporadic cases. Int J Med Sci 5: 197–202.
  3. 3. Gentilini O, Botteri E, Rotmensz N, Toesca A, De Oliveira H, et al. (2010) Breast-conserving surgery in 201 very young patients (<35 years). Breast 19: 55–58.
  4. 4. Cancello G, Maisonneuve P, Rotmensz N, Viale G, Mastropasqua MG, et al. (2010) Prognosis and adjuvant treatment effects in selected breast cancer subtypes of very young women (<35 years) with operable breast cancer. Ann Oncol 21: 1974–1981.
  5. 5. Peng R, Wang S, Shi Y, Liu D, Teng X, et al. (2011) Patients 35 years old or younger with operable breast cancer are more at risk for relapse and survival: a retrospective matched case-control study. Breast 20: 568–573.
  6. 6. Ortega Jacome GP, Koifman RJ, Rego Monteiro GT, Koifman S (2010) Environmental exposure and breast cancer among young women in Rio de Janeiro, Brazil. J Toxicol Environ Health A 73: 858–865.
  7. 7. El Saghir NS, Seoud M, Khalil MK, Charafeddine M, Salem ZK, et al. (2006) Effects of young age at presentation on survival in breast cancer. BMC Cancer 6: 194.
  8. 8. Anders CK, Hsu DS, Broadwater G, Acharya CR, Foekens JA, et al. (2008) Young age at diagnosis correlates with worse prognosis and defines a subset of breast cancers with shared patterns of gene expression. J Clin Oncol 26: 3324–3330.
  9. 9. Han W, Kang SY (2010) Society KBC (2010) Relationship between age at diagnosis and outcome of premenopausal breast cancer: age less than 35 years is a reasonable cut-off for defining young age-onset breast cancer. Breast Cancer Res Treat 119: 193–200.
  10. 10. Sidoni A, Cavaliere A, Bellezza G, Scheibel M, Bucciarelli E (2003) Breast cancer in young women: clinicopathological features and biological specificity. Breast 12: 247–250.
  11. 11. Fernandopulle SM, Cher-Siangang P, Tan PH (2006) Breast carcinoma in women 35 years and younger: a pathological study. Pathology 38: 219–222.
  12. 12. Azim HA Jr, Michiels S, Bedard PL, Singhal SK, Criscitiello C, et al. (2012) Elucidating prognosis and biology of breast cancer arising in young women using gene expression profiling. Clin Cancer Res. 1 18(5): 1341–51.
  13. 13. Colleoni M, Rotmensz N, Robertson C, Orlando L, Viale G, et al. (2002) Very young women (<35 years) with operable breast cancer: features of disease at presentation. Ann Oncol 13: 273–279.
  14. 14. Perou CM, Sørlie T, Eisen MB, van de Rijn M, Jeffrey SS, et al. (2000) Molecular portraits of human breast tumours. Nature 406: 747–752.
  15. 15. Anders CK, Deal AM, Miller CR, Khorram C, Meng H, et al. (2011) The prognostic contribution of clinical breast cancer subtype, age, and race among patients with breast cancer brain metastases. Cancer 117: 1602–1611.
  16. 16. Lalloo F, Varley J, Moran A, Ellis D, O’dair L, et al. (2006) BRCA1, BRCA2 and TP53 mutations in very early-onset breast cancer with associated risks to relatives. Eur J Cancer 42: 1143–1150.
  17. 17. Fackenthal JD, Olopade OI (2007) Breast cancer risk associated with BRCA1 and BRCA2 in diverse populations. Nat Rev Cancer 7: 937–948.
  18. 18. Haffty BG, Choi DH, Goyal S, Silber A, Ranieri K, et al. (2009) Breast cancer in young women (YBC): prevalence of BRCA1/2 mutations and risk of secondary malignancies across diverse racial groups. Ann Oncol 20: 1653–1659.
  19. 19. Walsh T, King MC (2007) Ten genes for inherited breast cancer. Cancer Cell 11: 103–105.
  20. 20. Lakhani SR, Gusterson BA, Jacquemier J, Sloane JP, Anderson TJ, et al. (2000) The pathology of familial breast cancer: histological features of cancers in families not attributable to mutations in BRCA1 or BRCA2. Clin Cancer Res 6: 782–789.
  21. 21. Palacios J, Honrado E, Osorio A, Cazorla A, Sarrió D, et al. (2005) Phenotypic characterization of BRCA1 and BRCA2 tumors based in a tissue microarray study with 37 immunohistochemical markers. Breast Cancer Res Treat 90: 5–14.
  22. 22. Brekelmans CT, Tilanus-Linthorst MM, Seynaeve C, vd Ouweland A, Menke-Pluymers MB, et al. (2007) Tumour characteristics, survival and prognostic factors of hereditary breast cancer from BRCA2-, BRCA1- and non-BRCA1/2 families as compared to sporadic breast cancer cases. Eur J Cancer 43: 867–876.
  23. 23. Hedenfalk I, Ringner M, Ben-Dor A, Yakhini Z, Chen Y, et al. (2003) Molecular classification of familial non-BRCA1/BRCA2 breast cancer. Proc Natl Acad Sci U S A 100: 2532–2537.
  24. 24. Jönsson G, Naylor TL, Vallon-Christersson J, Staaf J, Huang J, et al. (2005) Distinct genomic profiles in hereditary breast tumors identified by array-based comparative genomic hybridization. Cancer Res 65: 7612–7621.
  25. 25. Krepischi AC, Achatz MI, Santos EM, Costa SS, Lisboa BC, et al. (2012) Germline DNA copy number variation in familial and early-onset breast cancer. Breast Cancer Res 14: R24.
  26. 26. Carey LA, Perou CM, Livasy CA, Dressler LG, Cowan D, et al. (2006) Race, breast cancer subtypes, and survival in the Carolina Breast Cancer Study. JAMA 295: 2492–2502.
  27. 27. Oldenburg RA, Kroeze-Jansema K, Meijers-Heijboer H, van Asperen CJ, Hoogerbrugge N, et al. (2006) Characterization of familial non-BRCA1/2 breast tumors by loss of heterozygosity and immunophenotyping. Clin Cancer Res 12: 1693–1700.
  28. 28. Dutra MC, Rezende MA, de Andrade VP, Soares FA, Ribeiro MV, et al. (2009) [Immunophenotype and evolution of breast carcinomas: a comparison between very young and postmenopausal women]. Rev Bras Ginecol Obstet 31: 54–60.
  29. 29. Bacchi LM, Corpa M, Santos PP, Bacchi CE, Carvalho FM (2010) Estrogen receptor-positive breast carcinomas in younger women are different from those of older women: a pathological and immunohistochemical study. Breast 19: 137–141.
  30. 30. Carvalho LV, Pereira EM, Frappart L, Boniol M, Bernardo WM, et al. (2010) Molecular characterization of breast cancer in young Brazilian women. Rev Assoc Med Bras 56: 278–287.
  31. 31. Lourenço JJ VF, Bines J, Santos EM, Lasmar CAP, Costa CH, et al. (2004) BRCA1 mutations in Brazilian patients. Genetics and Molecular Biology 27: 500–504.
  32. 32. Figueiredo JC, Ennis M, Knight JA, McLaughlin JR, Hood N, et al. (2007) Influence of young age at diagnosis and family history of breast or ovarian cancer on breast cancer outcomes in a population-based cohort study. Breast Cancer Res Treat 105: 69–80.
  33. 33. Loizidou M, Marcou Y, Anastasiadou V, Newbold R, Hadjisavvas A, et al. (2007) Contribution of BRCA1 and BRCA2 germline mutations to the incidence of early-onset breast cancer in Cyprus. Clin Genet 71: 165–170.
  34. 34. Bonadona V, Sinilnikova OM, Chopin S, Antoniou AC, Mignotte H, et al. (2005) Contribution of BRCA1 and BRCA2 germ-line mutations to the incidence of breast cancer in young women: results from a prospective population-based study in France. Genes Chromosomes Cancer 43: 404–413.
  35. 35. Young SR, Pilarski RT, Donenberg T, Shapiro C, Hammond LS, et al. (2009) The prevalence of BRCA1 mutations among young women with triple-negative breast cancer. BMC Cancer 9: 86.
  36. 36. Bayraktar S, Gutierrez-Barrera AM, Liu D, Tasbas T, Akar U, et al. (2011) Outcome of triple-negative breast cancer in patients with or without deleterious BRCA mutations. Breast Cancer Res Treat 130: 145–153.
  37. 37. Robertson L, Hanson H, Seal S, Warren-Perry M, Hughes D, et al. (2012) BRCA1 testing should be offered to individuals with triple-negative breast cancer diagnosed below 50 years. Br J Cancer 106: 1234–1238.
  38. 38. Ginsburg OM, Akbari MR, Aziz Z, Young R, Lynch H, et al. (2009) The prevalence of germ-line TP53 mutations in women diagnosed with breast cancer before age 30. Fam Cancer 8: 563–567.
  39. 39. Garritano S, Gemignani F, Palmero EI, Olivier M, Martel-Planche G, et al. (2010) Detailed haplotype analysis at the TP53 locus in p.R337H mutation carriers in the population of Southern Brazil: evidence for a founder effect. Hum Mutat 31: 143–150.
  40. 40. Achatz MI, Olivier M, Le Calvez F, Martel-Planche G, Lopes A, et al. (2007) The TP53 mutation, R337H, is associated with Li-Fraumeni and Li-Fraumeni-like syndromes in Brazilian families. Cancer Lett 245: 96–102.
  41. 41. Bos PD, Zhang XH, Nadal C, Shu W, Gomis RR, et al. (2009) Genes that mediate breast cancer metastasis to the brain. Nature 459: 1005–1009.
  42. 42. Folgueira MA, Carraro DM, Brentani H, Patrão DF, Barbosa EM, et al. (2005) Gene expression profile associated with response to doxorubicin-based therapy in breast cancer. Clin Cancer Res 11: 7434–7443.
  43. 43. Castro NP, Osório CA, Torres C, Bastos EP, Mourão-Neto M, et al. (2008) Evidence that molecular changes in cells occur before morphological alterations during the progression of breast ductal carcinoma. Breast Cancer Res 10: R87.
  44. 44. Sørlie T, Perou CM, Tibshirani R, Aas T, Geisler S, et al. (2001) Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications. Proc Natl Acad Sci U S A 98: 10869–10874.
  45. 45. Parker JS, Mullins M, Cheang MC, Leung S, Voduc D, et al. (2009) Supervised risk predictor of breast cancer based on intrinsic subtypes. J Clin Oncol 27: 1160–1167.
  46. 46. Cabodi S, del Pilar Camacho-Leal M, Di Stefano P, Defilippi P (2010) Integrin signalling adaptors: not only figurants in the cancer story. Nat Rev Cancer 10: 858–870.
  47. 47. Damiano L, Di Stefano P, Camacho Leal MP, Barba M, Mainiero F, et al. (2010) p140Cap dual regulation of E-cadherin/EGFR cross-talk and Ras signaling in tumour cell scatter and proliferation. Oncogene 29(25): 3677–90.
  48. 48. Hayashi T, Arimura T, Itoh-Satoh M, Ueda K, Hohda S, et al. (2004) Tcap gene mutations in hypertrophic cardiomyopathy and dilated cardiomyopathy. J Am Coll Cardiol 44(11): 2192–20.
  49. 49. Gomes MC, Costa MM, Borojevic R, Monteiro AN, Vieira R, et al. (2007) Prevalence of BRCA1 and BRCA2 mutations in breast cancer patients from Brazil. Breast Cancer Res Treat 103: 349–353.
  50. 50. da Costa EC, Vargas FR, Moreira AS, Lourenço JJ, Caleffi M, et al. (2008) Founder effect of the BRCA1 5382insC mutation in Brazilian patients with hereditary breast ovary cancer syndrome. Cancer Genet Cytogenet 184: 62–66.