The Molecular Detection of Germline Mutations in the BRCA1 and BRCA2 Genes Associated with Breast and Ovarian Cancer in a Romanian Cohort of 616 Patients

The objective of this study was to identify and classify the spectrum of mutations found in the BRCA1 and BRCA2 genes associated with breast and ovarian cancer in female patients in Romania. Germline BRCA1 and BRCA2 mutations were investigated in a cohort of 616 female patients using NGS and/or MLPA methods followed by software-based data analysis and classification according to international guidelines. Out of the 616 female patients included in this study, we found that 482 patients (78.2%) did not have any mutation present in the two genes investigated; 69 patients (11.2%) had a BRCA1 mutation, 34 (5.5%) had a BRCA2 mutation, and 31 (5%) presented different type of mutations with uncertain clinical significance, moderate risk or a large mutation in the BRCA1 gene. Our investigation indicates the most common mutations in the BRCA1 and BRCA2 genes, associated with breast and ovarian cancer in the Romanian population. Our results also bring more data in support of the frequency of the c.5266 mutation in the BRCA1 gene, acknowledged in the literature as a founder mutation in Eastern Europe. We consider that the results of our study will provide necessary data regarding BRCA1 and BRCA2 mutations that would help to create a genetic database for the Romanian population.


Introduction
The autosomal dominant inherited cancer predisposition characterized by an increased risk of breast and ovarian cancers is called hereditary breast and ovarian cancer syndrome (HBOC).This type of familial/hereditary cancer represents approximately 10-15% of all breast cancer and 25% of ovarian cancer cases [1].Breast cancer has become the most commonly diagnosed cancer worldwide, overtaking lung cancer, with over 2.200.000new cases in 2020 in both sexes.In Romania, female breast cancer had 12.085 cases, with a cumulative risk (0-74) of 10.3, and the number of prevalent cases (5 years) was 45,263 in 2020 [2].Hereditary (familial) breast cancer represents only 10% of all breast cancer cases.Almost half of these are associated with mutations in BRCA1 (BReast CAncer gene 1) and BRCA2 (BReast CAncer gene 2) genes, while the other half are associated with mutations in different genes, such as PALB2 (partner and localizer of BRCA2), CHEK2 (checkpoint kinase 2), and PTEN (phosphatase and tensin homolog).Germline mutations in the BRCA1 and BRCA2 genes are reported in about 15-20% of all triple-negative breast cancers [3,4].Ovarian cancer, despite being only eighth place in the number of new cases in 2020 with 313.959, is a silent public health concern due to its high mortality rate.In Romania, ovarian cancer had 1.909 cases with a cumulative risk (0-74) of 1.7, and the number of prevalent cases (5 years) was 5.302 in 2020 [2].The most common subtype of epithelial ovarian cancer is the high-grade serous ovarian cancer (HGSOC) subtype, which is mostly associated with BRCA1/2 mutations [5][6][7] and, in some cases, with mutations in the RAD51C (RAD51 Paralog C), RAD51D (RAD51 Paralog D), and BRIP1 (BRCA1 Interacting Helicase 1) genes [8].Patients with this subtype of ovarian cancer who are negative for somatic mutations of the BRCA1 and BRCA2 genes will be subsequently tested for HRD (Homologous Recombination Deficiency) score evaluation.If the HRD score is positive, these patients could benefit from targeted therapies, such as PARP (poly (ADP-ribose) polymerases) inhibitors (PARPi) [9,10].Some of the major risk factors for developing breast or ovarian cancer consist of a family history of different types of cancer, the use of estrogen and progesterone replacement therapies, advanced age at first birth, exposure to radiation or other toxic environmental factors, and an unbalanced diet and lack of physical activity [11][12][13][14].
Carriers of BRCA1 mutations have an estimated lifetime risk of developing breast cancer of 55-72%, a 39-44% risk of developing ovarian cancer, a 21-29% risk for males of developing prostate cancer, and a 1-2% risk for males of developing breast cancer, as well as a 1-3% chance of developing pancreatic cancer.These estimated lifetime risks may differ according to the location or type of mutation [15].Carriers of BRCA2 mutations have a 45-69% risk of developing breast cancer, an 11-17% risk of developing ovarian cancer, and a 27-60% risk for males of developing prostate cancer, as well as a higher BRCA1 risk of 6-8% for males of developing breast cancer and a 3-5% risk of developing pancreatic cancer [10,16,17].
These two types of cancers (female breast and ovarian cancer) are mostly curable when they are discovered in the early stages of development and when medical treatment is available and affordable.Even though these types of cancer are treatable, chemoresistance and recurrence are very frequent and significantly worsen the prognosis of the disease.Genetic testing of the BRCA1 and BRCA2 gene mutations is particularly recommended for individuals with familial susceptibility to cancer or who have a personal history of cancer.In Romania, many patients are diagnosed only in the late stages of breast and/or ovarian cancer development due to a low level of population screening and prophylactic programs (a lack of funding and/or poor advertisement) to inform people about cancer risk [16,[18][19][20][21][22].
Hopefully, to possibly add another tool in helping to reduce the diagnostic time for patients with breast and/or ovarian cancer, our study brings new information about the most frequent mutations of BRCA1 and BRCA2 genes associated with breast or ovarian cancer in the Romania population.Our data, along with those reported by other groups from different regions of Romania [23][24][25][26], could represent the start of the development of a national cancer-related database that would help improve the management of the patients.

Materials and Methods
Biological samples consisted of peripheral blood collected from 616 female patients (spanning from January 2019 to March 2022) diagnosed with breast and/or ovarian cancer who were referred to Personal Genetics, Bucharest, Romania for BRCA1/2 germline mutation status.The study was conducted in accordance with the Declaration of Helsinki and was approved by the Institutional Review Board (or Ethics Committee) of Personal Genetics (107/16 March 2022).The analyses were performed at NeoScreen Genetics Center, Greece, and all patients completed and signed a written informed consent/GDPR agreement.
Patient selection.For ovarian cancer, we included in our study 432 patients diagnosed with non-mucinous ovarian cancer, which represents one of the most common subtypes of ovarian cancer associated with BRCA 1/2 mutations [13,27].Of the 432 patients, 390 had the high-grade serous ovarian cancer subtype and the remaining 42 had no defined subtype.For breast cancer, we included 184 patients who have been diagnosed with triple-negative breast cancer, a subtype of breast cancer that is characterized by having no expression of an estrogen receptor (ER), a progesterone receptor (PR), or human epidermal growth factor receptor 2 (HER2).This subtype is highly aggressive, has a shorter survival rate and a higher likelihood of recurrence, and was considered due to a higher association with BRCA1 gene mutations [20,28,29].
Total genomic DNA isolation from blood samples was performed using the "Nu-cleoSpin Tissue kit" (Product Code 11982262; Macherey-Nagel, Thermo Fisher Scientific, Waltham, MA, USA), according to the manufacturer's protocol.
The DNA sample was subjected to point mutation detection analysis (base substitutions and short deletions/duplications of a few base pairs), which was conducted through massively parallel sequencing of the coding regions of the BRCA1 (NM_007294.4) and BRCA2 genes (NM_000059.3),including exon-intron junction regions.Sequencing was performed using the "Accel Amplicon™ BRCA1, BRCA2 panel" kit (Product code: AL-52048; Swift Biosciences, Ann Arbor, MI, USA) on a "Next Generation Sequencing" platform MiSeq (Illumina, San Diego, CA, USA).The detection of large genomic rearrangements (deletions/duplications of 1 or more exons) in the BRCA1 and BRCA2 genes was conducted by multiplex ligation-dependent probe amplification (MLPA) using the products P002-D1 and P045-D1 (MRC-Holland, Amsterdam, The Netherlands).Capillary electrophoresis was performed using the ABI3130xl genetic analyzer.All the SNP (single nucleotide polymorfism) mutations identified were confirmed ulteriorly through Sanger sequencing.
The analysis of the FASTQ files obtained from the massively parallel sequencing process on the MiSeq sequencing platform was performed with the following steps: Quality assessments of the reads before alignment, which involve the exclusion of nucleotides/reads with low quality; Alignment process performed using the BWM-MEM algorithm, thus generating the specific BAM files; Quality control post alignment process and coverage analysis; Detection of mutations (substitutions, small deletions, small insertions/duplication) based on the GATK (Genome Analysis Toolkit) algorithm and the characterization of the detected sequence variants using the following databases: ClinVar, OMIM (Online Mendelian Inheritance in Man), 1000GENOMES, GnomeAD (Genome Aggregation Database), Varsome, LOVD (Global Variome shared Leiden Open-source Variation Database), BIC, and INVITAE.If a missense mutation is detected, in silico analysis is performed using the dbNSFP program.The large deletions and/or large duplications were detected via bioinformatics analysis of raw data obtained by MLPA (Multiplex Ligation-dependent Probe Amplification) technology.MLPA data analysis was performed using Coffalyser Net software v.140721.1958(MRC Holland, Amsterdam, The Netherlands).

Results
Our results show that out of the 390 patients diagnosed with non-mucinous ovarian cancer, the high-grade serous ovarian cancer subtype, 300 had a negative result, 48 had a pathogenic BRCA1 mutation, 24 had a pathogenic BRCA2 mutation, two had a mutation in the BRCA1 gene with uncertain significance, six had a mutation in the BRCA2 gene with an uncertain significance VUS (variant with uncertain significance), five had the moderate risk variant c.9976A>T (p.Lys3326Ter) in the BRCA2 gene, and five had a large deletion in the BRCA1 gene (Table 1).Of the forty-two patients diagnosed with non-mucinous ovarian cancer with no defined subtype, thirty were negative, three had a pathogenic BRCA1 mutation, six had a pathogenic BRCA2 mutation, one had a VUS mutation in the BRCA2 gene, and two had the moderate risk variant c.9976A>T (p.Lys3326Ter) in the BRCA2 gene (Table 1).
From the 184 patients diagnosed with triple-negative breast cancer, 152 had a negative result, 18 had a pathogenic BRCA1 mutation, four had a pathogenic BRCA2 mutation, two had a VUS mutation in the BRCA1 gene, two had a VUS mutation in the BRCA2 gene, four had the moderate risk variant c.9976A>T (p.Lys3326Ter) in the BRCA2 gene, and two had a large deletion in the BRCA1 gene (Table 1).

Variants in the BRCA1 Gene
Regarding their clinical significance, in the BRCA1 gene, the mutations were classified as pathogenic mutations (base substitutions and short deletions/duplications of a few base pairs-nineteen variants) (Table 2), VUS mutations (four variants) (Table 3), and large deletions of different regions with pathogenic significance (three variants).The three variants that affect large regions of the BRCA1 gene were detected by MLPA, and they are heterozygous deletions of the upper region of exons 1 and 2, heterozygous deletion of exon 8, and heterozygous deletions of exons 21 and 22. * HGVS nomenclature (DNA variants are enumerated according to NCBI reference sequence; the *-* symbol in the frequency columns represents the fact that in those databases, we did not find frequencies for these variants "https://www.ncbi.nlm.nih.gov(accessed on 3 October 2023)"; A-patients with non-mucinous ovarian cancer, high-grade serous ovarian cancer subtype; B-patients with non-mucinous ovarian cancer with no defined subtype; C-triple-negative breast cancer.* HGVS nomenclature (DNA variants are enumerated according to NCBI reference sequence; the *-* symbol in the frequency columns represents the fact that in those databases, we did not find frequencies for these variants "https://www.ncbi.nlm.nih.gov(accessed on 3 October 2023)"; A-patients with non-mucinous ovarian cancer, high-grade serous ovarian cancer subtype; B-patients with non-mucinous ovarian cancer with no defined subtype; C-triple-negative breast cancer.
The pathogenic and VUS mutations in the BRCA1 gene found in our group study are represented across protein domains in Figure 1 using the cBio Cancer Genomics Portal [30][31][32].

Variants in the BRCA2 Gene
Regarding their clinical significance, in the BRCA2 gene, the mutations were classified as pathogenic mutations (base substitutions and short deletions/duplications of a few base pairs-twenty-three variants) (Table 4), VUS mutations (eight variants) (Table 5), and one moderate risk variant c.9976A>T (p.Lys3326Ter), which is assigned as benign in the ClinVar database.Despite not being a pathogenic mutation, it has been previously shown in the literature that patients carrying this variant may be good candidates for PARP inhibitor therapy due to BRCA2 protein dysfunction induced by the modification of this variant over the C-terminal region of the protein [33].

Variants in the BRCA2 Gene
Regarding their clinical significance, in the BRCA2 gene, the mutations were classified as pathogenic mutations (base substitutions and short deletions/duplications of a few base pairs-twenty-three variants) (Table 4), VUS mutations (eight variants) (Table 5), and one moderate risk variant c.9976A>T (p.Lys3326Ter), which is assigned as benign in the ClinVar database.Despite not being a pathogenic mutation, it has been previously shown in the literature that patients carrying this variant may be good candidates for PARP inhibitor therapy due to BRCA2 protein dysfunction induced by the modification of this variant over the C-terminal region of the protein [33].* HGVS nomenclature (DNA variants are enumerated according to NCBI reference sequence; the *-* symbol in the frequency columns represents the fact that in those databases, we did not find frequencies for these variants "https://www.ncbi.nlm.nih.gov(accessed on 3 October 2023)"; A-patients with non-mucinous ovarian cancer, high-grade serous ovarian cancer subtype; B-patients with non-mucinous ovarian cancer with no defined subtype; C-triple-negative breast cancer.* HGVS nomenclature (DNA variants are enumerated according to NCBI reference sequence; the *-* symbol in the frequency columns represents the fact that in those databases, we did not find frequencies for these variants "https://www.ncbi.nlm.nih.gov(accessed on 3 October 2023)"; A-patients with non-mucinous ovarian cancer, high-grade serous ovarian cancer subtype; B-patients with non-mucinous ovarian cancer with no defined subtype; C-triple-negative breast cancer.
The pathogenic and VUS mutations in the BRCA2 gene found in our group study are represented across protein domains in Figure 2 using the cBio Cancer Genomics Portal [30][31][32].

Discussion
Our study results identified three main pathogenic mutations in the BRCA1 gene: c.5266dup (twenty-five samples), c.3607C>T (ten samples), and c.843_846del (five samples).
The variant c.5266dup (p.Gln1756ProfsTer74) found in the BRCA1 gene results from the insertion/duplication of a single nucleotide G in the position 5266 in exon 19 and is associated with the SNP reference rs80357906.This variant, also known in the literature as c.5329dup (p.Gln1777ProfsTer74), 5382insC, or 5385insC, is considered to cause a frameshift mutation and has a molecular consequence (by changing the reading frame), leading to the appearance of a premature termination codon.This type of modification alters protein synthesis, leading to the formation of a truncated protein that has no Cterminal BRCT (BRCA1 C Terminus) domain or no protein result, causing a loss of function in the gene.The protein affected by this variant is unable to form nuclear foci, regardless of the higher concentration in the cytoplasm being directed to the nucleus by BARD1 [34].The heterozygous loss of function is a confirmed disease mechanism for the Hereditary Breast and Ovarian Cancer Syndrome.The ClinGen (Clinical Genome)-approved ENIGMA (Enhancing NeuroImaging Genetics through Meta-Analysis) panel classified this mutation as pathogenic in 2016 according to the American College of Medical Genetics (ACMG) criteria: PVS1, PP5, and PM2.This variant was shown to be a frequent founder mutation in the Ashkenazi Jewish population, and it has been reported in many individuals affected by breast or/and ovarian cancer all over the world, including Europe, Asia, America, and Africa, with a high prevalence in Poland and Eastern Europe, including Romania [35][36][37][38][39][40][41][42][43][44].This variant has also been observed in patients with pancreatic or prostate cancer [45,46].Carriers of this mutation were shown to have a higher risk of developing breast or ovarian cancer by the age of 70, and the risk for breast cancer is 67-89% and 22-42% for ovarian cancer [47][48][49].The variant was reported in the ClinVar database with a pathogenic clinical significance (Variation ID: 17677) and is listed in various international databases: 0.000034 (TOPMED), 0.000183 (GnomAD_exome), 0.00040 (ALFA), and 0.000156 (ExAC).The Varsome clinical database classified this mutation as pathogenic, having a score of pathogenicity of 17 points.
The nonsense variant c.3607C>T (p.Arg1203Ter) found in the BRCA1 gene results from the substitution of the nucleotide C to nucleotide T in the 3607 position in exon 10 and is associated with the SNP reference rs62625308.This variant, also known as c.3726C>T, is considered to cause a premature stop codon, therefore leading to a premature termination codon in position 1203 of the protein.The appearance of this stop codon determines the synthesis of a truncated protein or the absence of a protein.In the same

Discussion
Our study results identified three main pathogenic mutations in the BRCA1 gene: c.5266dup (twenty-five samples), c.3607C>T (ten samples), and c.843_846del (five samples).
The variant c.5266dup (p.Gln1756ProfsTer74) found in the BRCA1 gene results from the insertion/duplication of a single nucleotide G in the position 5266 in exon 19 and is associated with the SNP reference rs80357906.This variant, also known in the literature as c.5329dup (p.Gln1777ProfsTer74), 5382insC, or 5385insC, is considered to cause a frameshift mutation and has a molecular consequence (by changing the reading frame), leading to the appearance of a premature termination codon.This type of modification alters protein synthesis, leading to the formation of a truncated protein that has no C-terminal BRCT (BRCA1 C Terminus) domain or no protein result, causing a loss of function in the gene.The protein affected by this variant is unable to form nuclear foci, regardless of the higher concentration in the cytoplasm being directed to the nucleus by BARD1 [34].The heterozygous loss of function is a confirmed disease mechanism for the Hereditary Breast and Ovarian Cancer Syndrome.The ClinGen (Clinical Genome)-approved ENIGMA (Enhancing NeuroImaging Genetics through Meta-Analysis) panel classified this mutation as pathogenic in 2016 according to the American College of Medical Genetics (ACMG) criteria: PVS1, PP5, and PM2.This variant was shown to be a frequent founder mutation in the Ashkenazi Jewish population, and it has been reported in many individuals affected by breast or/and ovarian cancer all over the world, including Europe, Asia, America, and Africa, with a high prevalence in Poland and Eastern Europe, including Romania [35][36][37][38][39][40][41][42][43][44].This variant has also been observed in patients with pancreatic or prostate cancer [45,46].Carriers of this mutation were shown to have a higher risk of developing breast or ovarian cancer by the age of 70, and the risk for breast cancer is 67-89% and 22-42% for ovarian cancer [47][48][49].The variant was reported in the ClinVar database with a pathogenic clinical significance (Variation ID: 17677) and is listed in various international databases: 0.000034 (TOPMED), 0.000183 (GnomAD_exome), 0.00040 (ALFA), and 0.000156 (ExAC).The Varsome clinical database classified this mutation as pathogenic, having a score of pathogenicity of 17 points.
The nonsense variant c.3607C>T (p.Arg1203Ter) found in the BRCA1 gene results from the substitution of the nucleotide C to nucleotide T in the 3607 position in exon 10 and is associated with the SNP reference rs62625308.This variant, also known as c.3726C>T, is considered to cause a premature stop codon, therefore leading to a premature termination codon in position 1203 of the protein.The appearance of this stop codon determines the synthesis of a truncated protein or the absence of a protein.In the same way as the previous mutation, the heterozygous loss of function is a confirmed disease mechanism for the Hereditary Breast and Ovarian Cancer Syndrome.The ClinGen-approved ENIGMA panel classified this mutation as pathogenic in 2016, and the following ACMG criteria were applied: PVS1, PP5, and PM2.This variant has been reported in many individuals affected by breast and/or ovarian cancer of different cultural ethnicities [41,[50][51][52][53][54][55][56].It has been reported in ClinVar that this mutation has a pathogenic clinical significance (Variation ID: 17671) and it has the following frequencies in different databases: 0.000012 (GnomAD_exome), 0.000010 (ALFA), and 0.00001 (PAGE_STUDY).The Varsome clinical database classified this mutation as pathogenic, having a score of pathogenicity of 17 points.
The variant c.843_846del (p.Ser282TyrfsTer15) found in the BRCA1 gene results from the deletion of four nucleotides (CTCA) at the position 843 to 846 in exon 10 and is associated with the SNP reference rs80357919.This variant, also known as 962del4, is considered to cause a frameshift mutation, leading to a premature termination codon, which determines a truncated protein or the lack of a protein.The loss of function in BRCA1 protein is a common mechanism for breast and ovarian cancer.The mutation has been observed in many individuals affected by breast and/or ovarian cancer [57][58][59][60][61][62][63].The ENIGMA panel and 12 other submitters classified this mutation as pathogenic in 2014, and the following ACMG criteria were applied: PVS1, PP5, and PM2.It has been reported in ClinVar that this mutation has a pathogenic clinical significance (Variation ID: 17683), and its global frequency is not known.The Varsome clinical database classified this mutation as pathogenic, having a score of pathogenicity of 17 points.
In our study, we also identified four BRCA1 VUS mutations: c.4843G>A (two samples), c.3711A>G (two samples), c.2666C>T (one sample), and c.994C>T (one sample).For the c.944C>T mutation, only computational prediction data were found with results that were inconclusive regarding the impact of this variant on protein structure and function.For the c.4843G>A mutation, along with computational predictions that were inconclusive, functional studies using BRCA1-deficient mouse embryonic stem cells in PARP inhibitor sensitivity assays have shown that this variant does not impact BRCA1 function in transcription activation [64].For c.3711A>G and c.2666C>T mutations, the published data suggest that these alterations are not expected to disrupt BRCA1 protein function.All of these VUS mutations have been previously reported in individuals affected with breast and/or ovarian patients [65][66][67][68][69].
The most frequent pathogenic mutations in the BRCA2 gene that we identified in our data are c.4284dup (three samples), c.5379_5380ins (three samples), c.7180A>T (three samples), and c.9371A>T (three samples).
The variant c.4284dup (p.Gln1429SerfsTer9) found in the BRCA2 gene results from the insertion/duplication of a single nucleotide T in the position 4284 in exon 11 and is associated with SNP reference rs80359439.This mutation, also known as 4512dupT or 4510insT, has a frameshift molecular consequence (changing the reading frame), which means that instead of the Glutamine amino acid, it determines the formation of a premature termination codon that leads to the synthesis of a truncated protein or the degradation of the protein, thus altering the protein function.The variant was classified by the ClinGenapproved ENIGMA as a pathogenic mutation, in accordance with the following ACMG criteria: PVS1, PP5, and PM2.It has been observed in many patients with breast and/or ovarian cancer [60,[70][71][72][73][74].This mutation was also described in one patient with medulloblastoma and in another patient with prostate cancer and other malignancies [75,76].It has been reported in ClinVar that this mutation has a pathogenic clinical significance (Variation ID: 37892), and it has the following frequencies in different databases: 0.000015 (TOPMED) and 0.000004 (GnomAD_exome).The Varsome clinical database classified this mutation as pathogenic, having a score of pathogenicity of 17 points.
The variant c.5379_5380insTT (p.Val1794LeufsTer2) results from the insertion of two nucleotides T in positions 5379_5380 in exon 11 and is associated with SNP reference rs879255329.This is a frameshift type of mutation that alters the open reading frame, creating a premature stop-codon that disrupts the translation of two amino acids thereafter.This interruption results in a truncated protein or no protein at all.This variant is classified as pathogenic by ClinVar (Variation ID: 37892) and Varsome Clinical (having a score of pathogenicity of 17 points), and its global frequency is unknown.
The variant c.7180A>T (p.Arg2394Ter) results from the substitution of an A to T in the 7180 position in exon 13 and is associated with SNP reference rs80358946.This mutation, also known as 7408A>T, has a gained stop codon, which leads to the loss of function of the protein because of its truncated conformation.It has been reported in many affected individuals with breast and/or ovarian cancer [77][78][79][80][81].This variant is classified as pathogenic in ClinVar (Variation ID: 52279) and Varsome Clinical (having a score of pathogenicity of 17 points), and its global frequency in GnomAD is 0.000007.
The variant c.9371A>T (p.Asn3124Ile) results from the substitution of A to T in the 9371 position in exon 24 and is associated with SNP reference rs28897759.This variant is a missense mutation that leads to the replacement of the asparagine amino acid with isoleucine in the 3124 position.This modification affects not only the conformation of the DNA binding domain but also the functionality of the protein [82][83][84][85][86].The variant was described in individuals affected by breast and/or ovarian cancer [87][88][89][90] and was shown to be a frequently found mutation in patients from Germany, Poland, and Romania [35,41,87,91].This variant is classified as pathogenic in ClinVar (Variation ID: 38233) and Varsome Clinical (having a score of pathogenicity of 16 points) and has the following global frequencies: 0.000007 (gnomAD) and 0.000004 (TopMed).
For c.3070A>G, c.191C>T, c.442T>C, c.2396A>G, and c.8254A>T mutations, the studies show that the sequence alterations are not expected to disrupt BRCA2 protein function.The c.4412_4414del and c.4446_4451dup mutations do not have, to our knowledge, any data from experimental studies or algorithm predictions [92].All these variants have been previously reported in individuals with breast and/or ovarian cancer [58,[93][94][95][96].

Conclusions
Our investigation revealed the most common mutations in the BRCA1 and BRCA2 genes associated with breast and ovarian cancer.Our data also support that the c.5266dup mutation in the BRCA1 gene could be considered a founder mutation in Eastern Europe, as previously described in the literature.Because of the recent studies that show the implications of mutations from other predisposing genes (e.g., PALB2, CHEK2, ATM, RAD51C, RAD51D, BRIP1, etc.) in these types of cancers and also the association of the BRCA1 and BRCA2 mutations in other types of cancers, it is necessary to extend the panel of genes tested and the types of cancer (e.g., pancreatic cancer, prostate cancer).We consider that the results of our study provide data regarding BRCA1 and BRCA2 mutations to create the genetic database needed for the Romanian population.

Table 1 .
Clinical and demographic characteristics of the patients included in our group study.

Table 2 .
The pathogenic mutations in the BRCA1 gene found in our group study.

Table 3 .
The mutations in the BRCA1 gene with unknown significance (VUS) found in our group study.

Table 4 .
The pathogenic mutations in the BRCA2 gene found in our group study.

Table 4 .
The pathogenic mutations in the BRCA2 gene found in our group study.

Table 5 .
The mutations in the BRCA2 gene with unknown significance (VUS) found in our group study.