We conducted a study in which we analyzed BRCA1/2 point variants and copy number alterations in 37 TNBC and 28 HGSC patients. The main goal of our study was to determine the prevalence of BRCA1/2 somatic variants and copy number alterations in both malignancies and whether they could represent a significant molecular marker.

Of the 65 successfully sequenced patients in our cohort, we were able to identify five-point variants in six different patients, three in BRCA1 gene and two in BRCA2 gene. Regarding copy number alterations we detected one copy number loss in BRCA1 gene and one copy number gain in BRCA2 gene. The genetic screening of BRCA1/2 genes using these patients genomic DNA indicated that five harbored a germline BRCA1/2 genetic alteration. While three harbored a somatic BRCA1/2 genetic alteration (Table 3).

The first variant reported in our study is c.66_67delAG (p.Glu23fs). It was detected in two different patients one was diagnosed with TNBC and the second with HGSC. According to Clinvar database, c.66_67delAG (p.Glu23fs) is a pathogenic variant of BRCA1 gene, caused by the deletion of two nucleotides, Adenine and Guanine at exon 2 of the gene. At the protein level, this deletion provokes the substitution of Glutamic acid by Valine at codon 23, which leads to a premature stop codon at position 16 of the new reading frame [10]. This pathogenic variant was reported by Struewing et al. as a founder mutation of the Ashkenazi Jewish population [11]. Also, it has been reported in many other populations worldwide and in the Moroccan population as well [12, 13].

The second variant reported in our study is c.2494_2495delCCinsTT (p.Pro832Leu). It was detected in both the genomic and the FFPE DNA of a TNBC patient. This variant is due to the deletion of two cytosine nucleotides and the insertion of two thymine nucleotides at exon 10 of BRCA1 gene. At the protein level, this variant is expressed by the substitution of proline amino acid by leucine. To the best of our knowledge, this variant has never been reported before in any database or research study.

The third variant reported in our study is c.4412delG (p.Gly1471fs). It was detected only in the FFPE DNA of a HGSC patient. According to Clinvar database, this variant is a pathogenic mutation of BRCA1 gene that results from the deletion of a guanine nucleotide at exon 14 of BRCA1 gene. At the protein level, it causes the substitution of a Glycine amino acid by Alanine at codon 1471 which introduces a premature stop codon at position 34 of the new reading frame. This variant was reported before by Winter et al. as a somatic mutation in a breast cancer case [14].

The fourth variant reported in our study is c.643delG (p.Glu215Lys). It was detected in both the genomic and the FFPE DNA of a TNBC patient. This variant is due to the deletion of guanine nucleotide at exon 8 of BRCA2 gene. At the protein level, it causes the substitution of glutamic acid by lysine, which creates a premature stop codon at position 658 of the new reading frame. To the best of our knowledge, this variant is not reported in any database or research study.

The last variant reported in our study is c.7632_7633delCG. It was detected only in the FFPE DNA of a HGSC patient. This variant is due to the deletion of two nucleotides; guanine and cytosine at exon 16 of BRCA2 gene. At the protein level, it causes the substitution of valine amino acid by phenylalanine at position 2545, which provokes a premature stop codon at position 2547 of the new reading frame. To the best of our knowledge, this variant has never been reported before.

Regarding copy number alterations, we detected one copy number loss in BRCA1 gene 17q21.31(41249158–41,251,946)× 1, The latter was detected in both the genomic and the FFPE DNA of the patient. BRCA1 gene exon deletions have been reported before in HBOC cases [15, 16]. Moreover, Pan et al., have reported BRCA1 gene exon 7–8 deletion, in hereditary ovarian cancer cases [17].

We also detected one copy number gain 13q13.1(32930545–32,930,833)× 3 in BRCA2 gene. This copy number alteration was not detected in the patient’s genomic DNA. BRCA2 exon 1–2 copy number duplications have been reported before in HBOC cases [18]. However, exon 15 somatic duplication in breast or ovarian cancer has not been reported before in any database or research study.

In our study, from the 37 successfully sequenced TNBC patients, we didn’t detect any somatic point variants or copy number alterations in both genes. Our study results line up with those reported by several research studies worldwide [19, 20] Nevertheless, other research studies reported somatic point variants in non-familial TNBC cases with a prevalence between 3 to 5% [21, 22]. On the other hand, from the 28 successfully sequenced HGSC patients, we detected two somatic pathogenic variants in BRCA2 gene (7%) and one somatic copy number gain also in BRCA2 gene (3.5%). Our findings are consistent with most studies worldwide, which reported BRCA1/2 somatic pathogenic variants in non-familial HGSC, with a prevalence of 5 to 8% [23,24,25]. However, concerning copy number alterations as far as we know most studies worldwide perform a comprehensive CNA analysis in non-familial TNBC and HGSC cases. Unfortunately, we couldn’t find a study that reported BRCA2 somatic copy number gain in HGSC [26, 27].

In our cohort four of our patients out of 65 (6%) harbored a BRCA1/2 pathogenic germline alteration, we concluded that these cases suffer from hereditary breast and ovarian cancer. Even though our selection criteria aimed to avoid HBOC patients. Several studies worldwide reported variable frequencies of BRCA1/2 germline variants in unselected TNBC and HGSC cases [28, 29]. All four patients that harbored BRCA1/2 germline alterations in our cohort confirmed the absence of cancer history in their families. Unfortunately, we could not analyze the genetic profile of their descendent parents. As a result, we could not confirm the hereditary origin of these genetic alterations. As far as we are concerned, they could be de novo alterations, which has been reported before by Kim De Leeneer et al. [30]. Moreover, we cannot neglect the fact that a respectable percentage of breast and ovarian cancer patients, with inherited BRCA1/2 genetic alterations, do not have a clear family history, due to a small family structure, the predominance of males in the family, and paternal inheritance [24]. This can only indicate that BRCA1/2 genetic testing should be considered for TNBC and HGSC patients even with the absence of hereditary breast and ovarian cancer selection criteria.

In the present study, we detected three point variants and one copy number gain that has never been reported before; c.2494_2495delCCinsTT (p.Pro832Leu), c.643delG (p.Glu215Lys), c.7632_7633delCG and 13q13.1(32930545–32,930,833)× 3. The Oncomine variant class pipeline has classified both variants c.643delG (p.Glu215Lys) and c.7632_7633delCG as truncating variants that causes a loss of normal protein function. While c.2494_2495delCCinsTT (p.Pro832Leu) variant was not classified by the Oncomine variant class pipeline therefore, to determine the pathogenicity of this variant we used the in silico analysis results given by SIFT (TI = 0.36), PolyPhen (PISC = 0.07) and GRANTHAM (R = 98). These results demonstrate that this variant is still of unknow significance. Apropos the copy number gain 13q13.1(32930545–32,930,833)× 3 the latter was also not classified by the Oncomine pipeline. However, since BRCA2 gene is a tumor suppressor gene. Copy number gains are not regarded as a cancer driver genetic alterations [28]. Nonetheless, they are considered as up-regulating genetic alterations that, depending on their position in the gene, may cause a protein overexpression. BRCA2 overexpression has been reported before in research studies that analyzed the expression profile in breast cancer tumors [31, 32]. More studies should be conducted regarding BRCA1/2 copy number variation and their phenotypical effects.

Both variants c.643delG (p.Glu215Lys) and c.66_67delAG (p.Glu23fs). Were detected in a homozygote state in the patient’s FFPE DNA. The second variant was detected in the patient genomic DNA as well in a heterozygote state. The fact that the latter is a germline pathogenic variant that was present in the patient’s FFPE DNA in a homozygote state, encourages us to support the two genetic hit theory proposed by Knudson et al. [33]. Nevertheless, three other patients in our cohort presented a germline BRCA1/2 pathogenic variant and they didn’t present a loss of heterozygosis or compound mutations in BRCA1/2 genes in their FFPE DNA. The genetic cancerogenesis process of hereditary breast and ovarian cancer is still not well established but many scientific studies have indicated that these tumors need an additional somatic mutation in other tumor suppress genes or oncogenes [34, 35].

To the best of our knowledge, this is the first study to analyze BRCA1/2 point variants and copy number alterations in the FFPE DNA of Moroccan TNBC and HGSC patients. Effectively, we were able to report two somatic point variants and one somatic copy number gain in HGSC patients. Moreover, we were able to identify three germline point variants that to the best of our knowledge were never reported before. This study offers information of clinical importance regarding BRCA1/2 genetic screening in non-familial cases of TNBC and HGSC. Nonetheless, the small number of cases in our cohort and the absence of information concerning the methylation profile of both genes are two limitations of our study, that should be surpassed in other future Moroccan studies dedicated to the implication of BRCA1/2 genes in breast and ovarian cancer.