Screening of BRCA1 /2 Mutations Using Direct Sequencing in Indonesian Familial Breast Cancer Cases

Breast cancer has emerged as the most prevalent cancer among women worldwide, including in Indonesia. The contribution of genes associated with high-risk breast-ovarian cancers, BRCA1 and BRCA2 , in the Indonesian population is relatively unknown. We have characterized family history of patients with moderate- to high-risk of breast cancer predisposition in 26 unrelated cases from Indonesia for BRCA1 /2 mutation analyses using direct sequencing. Known deleterious mutations were not found in either BRCA1 or BRCA2 genes. Seven variants in BRCA2 were documented in 10 of 26 patients (38%). All variants were categorized as unclassified (VUSs). Two synonymous variants, c.3623A>G and c.4035T>C, were found in 5 patients. One variant, c4600T>C, was found in a 38 year old woman with a family history of breast cancer. We have found 4 novel variants in BRCA2 gene including c.6718C>G, c.3281A>G, c.10176C>G, and c4490T>C in 4 unrelated patients, all of them having a positive family history of breast cancer. In accordance to other studies in Asian population, our study showed more frequent variants in BRCA2 compared to BRCA1 . Further studies involving larger numbers of hereditary breast cancer patients are required to reveal contribution of BRCA1 /2 mutations and/or other predisposing genes among familial breast cancer patients in Indonesia.


Introduction
Breast cancer is the most common cancer diagnosed among women worldwide including in Indonesia and the incidence is continuously increasing. (Akhsan and Aryandono, 2010;Jemal et al., 2011;Siegel et al., 2014) Although the mortality rate due to breast cancer improves during the past decades, social and economical burdens to the family and health system remain as the major problem in Indonesia. Especially in family with positive history of breast-ovarian cancer, other family members often feel worry for the higher risk to develop breast cancer (Purnomosari et al., 2007;Akhsan and Aryandono, 2010;Jemal et al. 2011).
Majority of breast cancer cases are sporadic without any familial predisposition. Around 5%-10% are diagnosed as hereditary cancer due to mutations in a single dominantly acting gene. However, only 25%-50% of hereditary breast cancers can be explained by mutations in the two most important breast cancer genes, BRCA1 and BRCA2 (Kobayashi et al. 2013). Hereditary breast cancer types are suggested in individuals with early onset of breast cancer (less than 40 years old), multiple primary cancers in a patient, clustering of breast-ovarian cancer RESEARCH ARTICLE

Screening of BRCA1/2 Mutations Using Direct Sequencing in Indonesian Familial Breast Cancer Cases
Sumadi Lukman Anwar 1 *, Samuel J Haryono 1,2,3 *, Teguh Aryandono 1 , I Gusti Bagus Datasena 2 within family, bilateral breast cancer, and first degree relatives of mutation carriers. Familial breast cancers have distinct morphology, response to treatment, and clinical outcome. BRCA1-associated breast cancers are predominantly triple-negative, basal-like type, and high grade. Family history and pedigree construction remain the most common approach to initially recognize hereditary/ familial breast cancer. Recently, relying on family history is relatively difficult because of small size of modern family and indecisive family history records (Daly et al., 2010;Gage et al., 2012).
Germline mutations in BRCA1 (MIM#113705; 17q chromosome) and BRCA2 (MIM#600185; 13q chromosome) genes are responsible for 25-50% of hereditary breast cancer. Women with BRCA1/2 mutations have an increased risk for breast cancer, ovarian cancer, fallopian tube cancer and other types of cancer. The risk for BRCA1-mutation carriers to develop breast cancer and ovarian cancer by age of 70 is 65-80% and 37-62%, respectively. While women with BRCA2 mutations have 45-85% and 11-23% lifetime risks to develop breast and ovarian cancer, respectively. The penetrance depends significantly on type of mutations as well as exogenous factors (Janavicius et al., 2010 ;Narod, 2010). In total 1,781 distinct variants in BRCA1 gene and 2,000 in BRCA2 gene have been documented in the Breast Cancer Information Core (BIC) database (January 2016). (The Breast Cancer Information Core Database., 2012.) BRCA1/2 proteins function as tumor suppressor and play an important role in the maintenance of genomic stability, DNA damage repair, cell cycle checkpoint, and transcriptional regulation (Roy et al., 2011;Kim et al., 2012).
In developed countries, genetic counseling and genetic testing for BRCA1 and BRCA2 are included in the standard of care for women at high-risk for breast-ovarian cancers. Genetic testing for BRCA1/2 is not yet commonly available in developing countries including in Indonesia due to limited health funding and limited information about prevalence of mutations of these genes Kobayashi et al., 2013). Until now, only one study has been addressed to detect mutations in BRCA1/2 genes in Indonesia population. To reveal prevalence of BRCA mutations in our population, we performed direct sequencing in 26 unrelated Indonesian patients with high risk of breast cancer predisposition for BRCA1/2 mutation analysis.

Ethics clearance
All patients have provided written inform consent to participate in this study and permit that the DNA samples would be genetically analyzed. This study has been approved by the ethical committees of The National Cancer Center of Dharmais Hospital Jakarta, Indonesia.

Patients and selection criteria
A total of 26 unrelated breast cancer patients were recruited from The National Cancer Center of Dharmais Hospital Jakarta, Indonesia. Breast cancer patients with moderate to high risk of a hereditary predisposition were recruited with the following criteria: (1) early-onset breast cancer (≤40 years), with or without a family history of breast or ovarian cancer in first-or second-degree relative; (2) younger than 50 years of age at diagnosis of breast cancer together with at least one first-or second-degree relative having breast or ovarian cancer at any age; or (3) bilateral breast cancer; or (4) two or more first-or second-degree relatives with breast and/or ovarian cancer, regardless of age. The clinicopathological characteristics of 26 breast cancer patients recruited in this study were summarized in Table 1. The eligible patients were then proposed for genetic counseling prior to consent. Collections of blood and tumor tissue samples were performed after inform consent from the selected patients. Patients with positive BRCA mutation testing were further informed for possible clinical surveillance and were offered to recruit their first-degree relatives for BRCA genetic testing.

DNA extraction
BRCA1 and BRCA2 mutation analysis was performed using genomic DNA extracted from 150 µl peripheral blood sample (for each patient) with DNA isolation kit (Roche, Germany). A ratio of absorbance at 260 nm and 280 nm of the purified DNAs between 1.7 -1.8 is generally accepted for PCR and subsequent direct sequencing. Agarose gel electrophoresis was used to check the DNA integrity.

Preparation for PCR
For DNA amplification, primers that have been previously used by De Leeneer et al. (2012) and human genome variation society (HGVS) were used in this study. A total of 39 and 52 PCR reactions were performed for amplification of BRCA1 and BRCA2 genes, respectively. Polymerase chain reaction was conducted in 25 µl PCR cocktail containing: 62.5 ng genomic DNA, 1X PCR buffer MgCl2 (Roche, Germany), 0.2 mM dNTP mx (Roche, Germany), 10 pmol of each primer (FirstBase, Singapore), 1.25 IU of Taq DNA polymerase (Roche, Germany). All reactions were performed in a thermal cycler (MyCycler, Biorad) with a program: pre-denaturation (95 o C in 5 min); denaturation, annealing, and elongation in 40 cycles (95 o C in 20 sec, 59-66 o C in 30 sec, 72 o C in 40 sec, respectively); and final elongation (72 o C in 5 min). Polymerase chain reaction was not conducted at the exon 9 of BRCA1 because we could not obtain optimal temperature for the annealing stage. In addition, exon 21 and 26 of BRCA2 were also not also amplified because the regions were not eligible for PCR amplification (short exon). PCR products were checked in agarose gel DNA electrophoresis with 1X Tris-Acetate-EDTA buffer system (2.5% agarose gel) prior to sequencing.

Direct sequencing of PCR products
PCR products were then purified using high pure PCR product purification kit (Roche, Germany). Precipitation of purified PCR products was carried out using DYEnamic ET terminator kit manual (GE Healthcare, USA). Direct sequencing was performed in GE MegaBACE 750 DNA analyzer/capillary electrophoresis (GE Healthcare, USA) following the manufacturer's instructions. DNA sequences were then analyzed using BioEdit Alignment software (Carlsbad, USA) and reviewed manually by two independent persons.

Mutation database
The genetic variants identified in our patients were compared with online published database from the Breast Information Core Database of National Institute of Health (BIC). Any findings of BRCA1/2 mutations were documented and compared.

Family history and clinicopathological characteristics
Detailed pedigree assessment and construction were formulated from deep anamnesis from each patient. A standard epidemiological questionnaire, including a detailed personal and family history, was also collected. Epidemiological questionnaire provided some information related to age at breast cancer diagnosis; marital status; age of menarche; other cancer diagnosed in the patient; a family history of breast, ovarian, and other cancers in firstor second-degree relatives; history of breast or ovarian surgery. Patient's medical records were collected and reviewed to complement the personal and familial history, clinicopathological data, and follow-up of treatment. Tumor stage was determined based on the seven edition of the American Joint Committee on Cancer staging system. Expression of estrogen receptor (ER), progesterone receptor (PR), and HER2 from immune-staining of the tumor tissues were documented to sub-classify patients.

Patient characteristics
A total of 26 patients with moderate and high risk of a hereditary breast cancer predisposition with a mean age of 44.8 years old (range 29-58 years old) were recruited to this study. All of them were referred and treated at the National Cancer Center of Dharmais Hospital, Jakarta,-Indonesia. All patients had a family history of breast cancer. No patient was identified with a family history of ovarian cancer in the first and second degree family.
Histological data showed that 76.9% patients had invasive ductal carcinoma. In our study, 69.2% patients were positive for estrogen-receptor and 7.6% were positive for progesterone-receptor. The clinicopathological data of these patients are summarized in Table 1.

BRCA1/2 sequence variants
Direct sequence analysis of all coding exons in BRCA1 and BRCA2 were performed in 26 familial breast cancer patients. Seven variants at the BRCA2 gene were identified. Deleterious BRCA1/2 gene mutations were not documented. In total, 10 (10/26, 38.5%) breast cancer patients had single nucleotide substitution in BRCA2 gene. Of these 10 patients, 5 patients (5/26, 19.2%) had synonymous mutation and 5 other patients (5/26, 19.2%) showed missense mutation ( Table 2). All variants documented in our study were classified as variants of uncertain significant (VUSs). According to the BIC, four out of the 10 VUSs in BRCA2 gene were novel substitutions which had not yet been published before. BRCA1 mutations or variants were not found in our study.
A variant found in BRCA2 exon 11 c.6718C>G led to amino acid glutamine to glutamic acid substitution. While alteration of BRCA2 c.3281A>G caused substitution of lysine to arginine. BRCA2 variant at c.4490T>C resulted in substitution of phenylalanine to serine. A single nucleotide substitution was identified at the exon 27 c.1017C>G and caused conversion of amino acid glutamine to glutamic acid. All 4 variants were novel mutations and were not registered before at the BIC ( Table 2).
The c.3623 A > G synonymous mutation was found in 4 unrelated breast cancer patients with diagnosis at the age of 34 years old, 37 years old, 42 years old, and 45 years old. All of these patients had first or second degree familial history of breast cancer in which her maternal aunt, mother, mother, mother and maternal cousin were respectively affected. In addition, 4 variants including c.4035 T > C (synonymous), c.6718 C > G (missense), c.3281 A > G (missense), c.4600 T > C (missense) were identified from respectively 34 years old, 46 years old, 52 years old, 38 years old women whose their mothers had

Discussion
Segregation analysis studies have shown that approximately 5,-10% breast and/or ovarian cancer cases are inherited with autosomal dominant pattern. Germline mutations of BRAC1 and BRCA2 genes account for the genetic predisposition and significantly increase the risk of breast and ovarian cancer. Individuals with BRCA1/2 mutations have probability 45-55% to develop breast cancer and 17-39% to develop ovarian cancer at the age of 70 years. (Chen et al., 2007;Ripperger et al., 2009) BRCA1/2 mutations are more frequent among women with first and second degree history of breast or ovarian cancer, a personal history of young age breast cancer, triple-negative subtype breast cancer (Chen et al., 2007;Daly et al., 2010;Hartman et al., 2012). The frequency of BRCA1/2 mutations varies greatly among different ethnic groups and countries (Hall et al., 2009). In developed nations, genetic cancer risk assessment guideline (including genetic testing) for inherited breast cancer susceptibility have become a standard for clinical management for high-risk patients (Narod et al., 2010). In developing countries including Indonesia, management of patients with high-risk for a predisposition of familial breast and ovarian cancer are not yet standardized. Study focusing on familial breast cancer and BRCA mutation spectrum in developing countries is often limited (Agarwal et al., 2009;Kim et al., 2013).
In our study, from a total of 26 breast cancer patients with hereditary predisposition, we did not find any deleterious BRCA1/2 mutation. Although deleteriuos mutations in BRCA1/2 genes were not found, we have documented variants at the BRCA2 gene in 10 patients. Five patients (19.2%) harbored synonymous mutation and the other 5 patients (19.2%) carried missense mutations. In the synonymous mutations, sequence alternations were identified without any amino acid substitution. Two synonymous mutations in the BRCA2, c.3623A>G and c.4035T>C, were detected in 4 and 1 breast cancer patients, respectively. All of them have family history of breast cancer and manifest as infiltrative ductal carcinomas.
Missense mutations, on the other hand, led to a single amino acid change. Both synonymous and missense mutations identified in our study were classified as variants of unknown significance (VUSs). We reported 4 novel BRCA2 variants including three variants were located at the exon: 11 c.6718C>G (p.Q2164E), c.3281A>G (p.K1018R), and c.4490T>C (p.F1421S) as well as a variant at the exon 27 c.10176C>G (p.Q3316E). To predict possible functional impact for the novel BRCA2 variants, we used in silico analysis using Align-GVGD and PANTHER (Brunham et al., 2005;Tavtigian et al., 2006). Align-GVDV calculates Grantham Variation (GV) and Grantham Deviation (GD). GV shows degree of biochemical variation among amino acid at a given position within multiple sequence alignments (MSA). GD predicts the distance between a missense substitution and range of variation observed at its position in the alignment. PANTHER calculates subPSEC scores that have been previously shown to be able to statistically differentiate Mendelian disease-associated missense mutations from random coding polymorphisms. The missense variant Q2164E has subPSEC score -2.07, Pdeleterious 0.28, GV=353.8, and GD=0. The second variant pK1018R has subPSEC score -1.38, Pdeleterious 0.16, GV=266, and GD=0. The pE3316Q variant has subPSEC score -1.9, Pdeleterious 0.25, GV=112, and GD=2.03 and the pF1421S variant has subPSEC score -1.8, Pdeleterious 0.24, GV=112, and GD=0. GV=0 represents in variant residue in the alignment, GV=60-65 indicates upper limit of conservative variation across species, and GV > 100 designates residue with slight functional constraint. GD=0 represents a missense variant within cross-species range of variation, GD=60-65 corresponds with upper limit of a conservative missense variant, and GD > 100 indicates radical substitution (Tavtigian et al. 2008). Using PANTHER, subPSEC scores have value ranging from 0 (neutral) to 10 (most likely as deleterious mutation). SubPSEC is able to be converted into Pdeleterious value as equation 1 as the most probable for detelerious effect (Brunham et al. 2005). Therefore, our novel identified  DOI:http://dx.doi.org/10.7314/APJCP.2016.17.4.1987 VUSs are not likely to affect BRCA2 protein functions.
Only one study has previously addressed to BRCA1/2 germline mutations in familial breast cancers among Indonesian population. This study focused on youngonset (<41 years-old) breast cancer patients using PCR-DGGE as pre-screening mutation analysis. Thirty three mutations at BRCA1/2 mostly VUSs were found among 120 early-onset Indonesian breast cancer patients (Purnomosari et al., 2007). Therefore, the contribution of BRCA1/2 gene mutations in familial breast cancer cases among Indonesian population is relatively uncovered. Among Asian population, most studies showed more frequent BRCA2 than BRCA1 mutations in familial breast-ovarian cancer patients Han et al., 2013). In multiethnic groups such as in Indonesia, prevalence of BRCA1/2 mutations is relatively difficult to be estimated. In addition, other genes or genetic modifiers might contribute to the familial breast cancer cases in our population (Bruin et al., 2012). Further studies with larger cohort addressing genomic BRCA1/2 rearrangement and mutation screening of gene functionally related to BRCA1/2 including PALB2, NBS1, CHEK2, ATM, and BRIP1 are warranted (Seong et al., 2009;Kang et al., 2010;Zhang et al., 2011;He et al., 2012).