The present study was based on residents of the city of Alexandria, located on the northern coast of Egypt. The majority of the included participants were educated, working women. A total of 11 families were included as follows: Six patients with FBC with a family history of BC (at least one first-degree relative affected before the age of 50 years) and 10 high-risk relatives. All the patients with FBC had a history of BC from the maternal side, apart from patient no. 4, who had a history of BC from both the maternal and paternal side. Additionally, there were five patients with SBC without a known family history of BC and six high-risk relatives. The clinicopathological characteristics of the patients are summarized in Table I. Patients were recruited between October 2018 and October 2019 from the Clinical Oncology Department, Faculty of Medicine and Medical Research Institute, Alexandria University (Alexandria, Egypt). There were no specific exclusion criteria, and all patients were included regardless of hormonal status and metastatic state. Patients and relatives signed a formal consent form to participate in the study. The present study was approved by the Ethical Committee, Faculty of Medicine, Alexandria University (approval no. 0304918).

For patients with both FBC and SBC, formalin-fixed paraffin-embedded (FFPE) tissue blocks were collected; the fixative used for those archived blocks was 10% buffered formalin overnight at room temperature for breast lumpectomy and modified radical mastectomy specimens at room temperature. The immunohistochemistry data were collected from the pathological reports of the cases, and the previously obtained slides were reviewed to confirm the correct interpretation. Following a pathological examination, DNA extraction was performed using a QIAamp DNA FFPE Tissue kit (cat. no. 56404; Qiagen GmbH). For relatives, genomic DNA was extracted from the peripheral blood using a PureLink™ Genomic DNA Mini kit (cat. no. K1820-00; Invitrogen; Thermo Fisher Scientific, Inc.). DNA was quantified using a Qubit™ 1X dsDNA HS (High Sensitivity) Assay kit Q33231 with a Qubit Fluorometer version 4 (Thermo Fisher Scientific, Inc.), according to the manufacturer's instructions.

A total of 10 ng DNA per sample was used for library preparation using the Ion AmpliSeq™ Library Kit Plus (cat. no. 4488990; Ion Torrent; Thermo Fisher Scientific, Inc.) and the Oncomine™ BRCA Research assay (cat. no. A32842; Ion Torrent; Thermo Fisher Scientific, Inc.), according to the manufacturer's instructions. Coding regions from BRAC1 and BRCA2 were amplified using 167 primer pairs divided into two pools. Each library was barcoded using the Ion Xpress™ Barcode Adapters 1–16 kit (cat. no. 4471250; Ion Torrent; Thermo Fisher Scientific, Inc.). An Ion Library TaqMan™ Quantitation kit (cat. no. 4468802; Ion Torrent; Thermo Fisher Scientific, Inc.) was used to quantify the libraries against a standard curve of serial dilutions of the E. coli DH10B Control Library provided (Thermo Fisher Scientific, Inc.). Barcoded libraries were pooled and diluted to a final concentration of 100 pM.

Template preparation by emulsion PCR on Ion Sphere Particles (ISPs) was performed using the Ion 520™ & Ion 530™ Kit-OT2 kit (cat. no. A27751) and the Ion OneTouch™ 2 System (Ion Torrent; Thermo Fisher Scientific, Inc.). The samples were loaded on Ion 520 Chip kit v2 (Ion Torrent; Thermo Fisher Scientific, Inc.). Sequencing primer and polymerase were added to the final enriched ISPs prior to loading onto an Ion 520 chip, according to the Ion 520 & Ion 530 kit-OT2 protocol.

Sequencing was performed using the Ion S5™ next generation sequencing system (Ion Torrent; Thermo Fisher Scientific, Inc.). Sequencing coverage for reporting were ≥20 unique reads (×20) for each base. Allelic frequency for somatic variants ranged from 0.05 to 1%, while for germline variants, the allele frequency was between 0.4 and 1%. Median coverage typically ranged between ×200 and ×300.

Sequencing data were processed using Torrent Suite software 5.12.1 (Ion Torrent; Thermo Fisher Scientific, Inc.) to analyze barcode reads; to align reads to the hg19 reference genome; and to generate run metrics, which include chip loading efficiency, total read counts and quality. Following statistical analysis, the annotation of single-nucleotide variants, insertions, deletions and splice site alterations, was performed using the Ion Reporter Server System (Oncomine BRCA Research Assay; Thermo Fisher Scientific, Inc.). The variants identified were further analyzed using Oncomine Reporter.

All mutations reported were confirmed in the Human Gene Mutation Database (HGMD; http://www.hgmd.cf.ac.uk/ac/index.php), Catalogue Of Somatic Mutations In Cancer (https://cancer.sanger.ac.uk/cosmic), the algorithms SIFT 6.2.1 (https://sift.bii.a-star.edu.sg/www/code.html) PolyPhen-2 (version 2) (http://genetics.bwh.harvard.edu/pph2/) and Align-GVGD (http://agvgd.hci.utah.edu/agvgd_input.php) and ClinVar (http://www.ncbi.nlm.nih.gov/clinvar/), and reviewed by the BRCA Exchange (https://brcaexchange.org). The mutation function and mutation classes were further investigated using the Consortium of Investigators of Modifiers of BRCA1/2 (CIMBA) (15).

The median age of patients with FBC and SBC was 35.5 and 66 years, respectively. Each patient had a history of unilateral BC, apart from FBC patient no. 6, who had a history of contralateral BC. With regards to first-degree relatives, no one had experienced a health issue, apart from the daughter of SBC patient no. 9, who had a history of fibroadenoma in the right breast (Tables I and II). The investigation of BRCA1 and BRCA2 variants in patients revealed 22 mutations, namely 9 in BRCA1 and 13 in BRCA2 (Table II).

All patients with FBC and SBC carried BRCA1 and BRCA2 mutations, apart from patients no. 6 and 11 who were carriers of BRCA2. Generally, no significant associations were detected between BRCA1 and BRCA2 mutations and patient age (data not shown). In total, ~82% of patients had infiltrating ductal carcinoma, 46% of patients had estrogen receptor (ER)- and progesterone receptor (PR)-positive BC of varying histological grades. In addition, three patients had triple-negative (ER, PR and Her2/neu) BC; two had FBC and one had SBC with a high histological grade (Table I). No significant association was observed between the hormonal status of the tumor, and BRCA1 and BRCA2 mutations (data not shown).

The analysis of pathogenic variants in patients revealed two variants in BRCA1 and five variants in BRCA2 with no preference to a specific exon (Table II). Pathogenic variants were mainly found in patients with FBC (five with FBC and three with SBC) (Table II). A pathogenic variant c.9089del BRCA2, located in exon 23 was shared by three patients with FBC (nos. 2, 3 and 6) with a median age of 34 years and who presented with early-onset BC: Two of them had triple-negative BC of high histological grade (III/IV) (Tables I and II). The variant c.9089del is a frameshift class 1 mutation that is predicted to encode a truncated non-functional protein (15). This variant was previously detected in Chinese patients with early-onset BC (16). Additionally, a missense pathogenic variant, c.8243G>A BRCA2, located in exon 18, was shared by FBC patient no. 5 and two patients with SBC (nos. 9 and 10) with medium to high-histological grades (II/III). Variant c.8243G>A is a class 2 mutation with a stable mutant protein (17). A class 2 missense pathogenic variant c.5053A>G BRCA1, located in exon 16, was found in FBC patient no. 3 and SBC patient no. 10 (grade II/III). This variant is linked to an increased risk of breast and ovarian cancer (18,19). Other pathogenic variants were detected, such as a class 1 frameshift c.5583_5584insA BRCA2 variant, located in exon 11 and observed in SBC patient no. 11 (grade II). Two intronic variants (BRCA2 c.7008-2A>T intron 13 with non-functional transcript, BRCA1 c.4097-1G>A intron 10) were identified in FBC patient no. 4 with a very high histological grade (III/IV). Both variants have been found to have splicing aberrations (20,21). In addition, one in-frame class 2 c.7976G>A BRCA2 located in exon 17 was found in FBC patient no. 5 (grade II). This variant has been previously reported in patients with BC and/or ovarian cancer (19,22).

Furthermore, as presented in Table II, four VUS were identified in four patients: Three variants in patients with FBC (two in BRCA2 and one in BRCA1) and one variant in BRCA2 in a patient with SBC. The variants were distributed among different exons: 11, 13, 20, 23 and intron 2. The BRCA exchange revealed that the synonymous substitution variant c.9087G>A had a low bioinformatic likelihood of resulting in a splicing aberration. Variant c.68-16delT has no existing functional assay data. The algorithms SIFT, PolyPhen-2 and Align-GVGD that predict the effect of missense changes on protein structure and function, all suggested that the variants c.4412G>T and c.6556T>A were likely to be tolerated.

One patient with SBC (patient no. 8) had a probable benign variant c.8503T>C BRCA2 in exon 20. This variant has previously been identified in two North African countries: Tunisia and Algeria (23).

Missense benign mutations with amino acid replacement were also identified: Six variants in BRCA1 and four variants in BRCA2. These benign variants were observed in both FBC and patients with SBC (Table II). Benign missense mutations were found in exon 10 (40%) and exon 11 (20%) followed by 10% each in exons 13, 14, 15 and 16. These benign variants located in exon 10 and exon 11 are considered ‘cold spots’ since they are tolerant to variations (24).

The investigation of BRCA1 and BRCA2 variants in first-degree relatives revealed three pathogenic variants, with all of these found in relatives of patients with FBC (Table III). Variant c.1278delA (GA>A) BRCA2 in exon 10 was detected in a daughter of patient no. 2, where the G allele is predicted to cause a frameshift deletion mutation with an amino acid change. Both relatives of FBC patient no. 4 had pathogenic variants: The mother had c.1960_1961delA, (CT>C, CTT) and the daughter had c.1224del, (CT>C). The last two variants were frameshift mutations located in BRCA1 exon 10. Variant c.1960_1961delA was also found in a sister of FBC patient no. 3. No pathogenic variants were identified in the relatives of patients with SBC.

Moreover, VUS, likely benign, or conflicting interpretation of pathogenicity variants were identified in 9/16 relatives: Six relatives with FBC and three with SBC. Three variants were detected in BRCA1 (exons 10 and 16) and four variants in BRCA2 (exons 10, 11 and intron 2). Notably, BRCA2 intron 2 c.68-16delT was observed in six at-risk relatives in four families (familial and sporadic). This variant was repeated in family no. 6 (patient and two relatives) and in both relatives of patient no. 4. The A allele in c.68-16delT (AT>A), is predicted to cause an unknown/non-coding mutation in the BRCA2 gene. This variant was previously selected as one of the specific deletions in BC microarray analyses (25). In addition, a daughter of FBC patient no. 1 had a likely benign BRCA1 exon 10 missense mutation. This locus is linked to adenocarcinoma in the Catalogue Of Somatic Mutations In Cancer (https://cancer.sanger.ac.uk/cosmic). Furthermore, one sister of FBC patient no. 3 had a BRCA1 exon 10 c.2733A>G synonymous variant that was previously detected only in patients with unilateral BC (24). One daughter of SBC patient no. 11 had a missense BRCA2 mutation c.5640T>G with a conflicting interpretation of pathogenicity that was previously identified as an unclassified variant in both contralateral and unilateral BC (26). Finally, both daughters of SBC patient no. 9 shared a BRCA2 exon 10 c.1011C>T synonymous variant. One of these daughters had a fibroadenoma in one breast and a history of pregnancy and lactation.

Seven out of 11 families had repeated benign BRCA missense mutations found in both patients and their relatives. Two benign mutations, c.3113A>G BRCA1 in exon 10 and c.4837A>G BRCA1 in exon 16, were detected in 8/16 relatives (50%). Variants c.3113A>G and c.4837A>G were previously reported as candidate founder mutations in an Iranian population (27).