Abstract
Introduction
Germline BRCA1/2 (gBRCA1/2) variants are strongly associated with hereditary cancers, and screening for these variants in high-risk populations is recommended for personalized management. This study aims to comprehensively characterize gBRCA1/2 variants in cancer and family screening cohorts from the Dubai Emirate, UAE.
Material and methods
A total of 443 patients with breast, ovarian, prostate and pancreatic cancer were tested for gBRCA1/2 variants from 2017 to 2022 using whole-gene sequencing, and data were analysed using variant interpretation and in-silico prediction tools. All BRCA1/2 variants were classified as P/LP or variants of uncertain significance (VUS) according to ACMG guidelines.
Results
In the cancer cohort, 38 out of 306 patients harboured gBRCA1/2 P/LP or VUS variants. Of these, 23 (7.5%) were classified as BRCA1/2 P/LP, while 15 (4.9%) were categorized as VUS. These variants were predominantly observed in estrogen receptor-positive/progesterone receptor-positive (ER + /PR +) and triple-negative breast cancer patients. Common BRCA1 P/LP variants included deletion frameshift variants (c.4065_4068del, c.68_69delAG, c.3228_3229delAG), an insertion frameshift variant (c.1140dup), and a nonsense variant (c.5251C > T). BRCA2 P/LP variants included a nonsense variant (c.5645C > A), a missense variant (c.7007G > A), and a deletion frameshift variant (c.2254_2257del). In the family screening cohort, 14 out of 137 samples harboured BRCA1/2 P/LP orVUS. Of these, five (3.6%) were classified as P/LP, while nine (6.6%) were VUS. Pathogenic BRCA1 variants included deletions (c.4065_4068del, c.3756_3759del) and a nonsense variant (c.5095C > T), while BRCA2 PVs included a deletion frameshift (c.771_775del) and a novel missense variant (c.8377G > A). In both cohorts, novel distinct variants were observed.
Conclusion
gBRCA1/2 variant prevalence in cancer and family screening cohorts can serve as beneficial personalized tool for management and treatment of cancer patients. Larger studies from other emirates of UAE will serve as a foundation for robust risk assessment and implementation of treatment and prevention strategies.
Supplementary Information
The online version contains supplementary material available at 10.1007/s00432-025-06188-9.
Keywords: BRCA1, BRCA2; Breast cancer, ovarian cancer; Prostate cancer; Pancreatic cancer; Prevalence; United Arab Emirates
Introduction
Cancer is a significant public health concern, with an estimated 19.96 million new cases and 9.74 million cancer deaths reported in 2022 (Sung et al. 2021). In UAE, cancer is the third-leading cause of death, accounting for 13.11% of all deaths (Shelpai 2019). Breast cancer (BC) is ranked first among cancers in women (36.7%) with majority of BC (21.5%) occurring at young age (between 30 and 40 years of age) (Al-Shamsi et al. 2023). In addition to BC, prostate, ovarian and pancreatic cancers have been reported with an incidence of 5.6%, 1.9% and 1.5% respectively indicating their significance in cancer statistics of UAE population (WHO IAfRoC.2024https 2024).
Hereditary breast and ovarian cancer (HBOC) are part of a broader category of diseases known as hereditary cancer syndromes, which also include conditions like Bloom syndrome, Fanconi anaemia, Nijmegen breakage syndrome, and Ataxia-telangiectasia. These syndromes are characterized by the inheritance of PVs that significantly increase the risk of developing various cancers. HBOC is associated with variations in the BReast CAncer gene (BRCA1 and BRCA2), that are crucial for DNA repair and maintaining genomic stability (Imyanitov et al. 2023). Individuals with variations in BRCA1 and BRCA2 genes carry a defective protein function with higher likelihood of developing breast and ovarian cancers, often at a younger age than the general population (Imyanitov et al. 2023). In addition to HBOC, presence of BRCA1 and BRCA2 variants have also been implicated in prostate and pancreatic cancers, with BRCA variant carriers having a 1.90- and 3.51-fold greater risk of developing prostrate and pancreatic cancers, respectively (Oh et al. 2019; Wong et al. 2020). Conversely, cancer patients tested positive for BRCA variants are eligible for targeted therapeutic options, such as poly-ADP ribose polymerase (PARP) inhibitors and platinum-based chemotherapies that specifically target BRCA mutant proteins for better patient outcomes (Farmer et al. 2005; Luo and Keyomarsi 2022).
A family history of cancer is a significant risk factor in predisposition to cancers. Therefore, screening of individuals for germline pathogenic variants (PVs) (with a family history of cancer) becomes an essential tool for the screening and management strategies. Therefore, routine screenings are part of mandatory guidelines in a clinical setting (Network and (NCCN). 2025).
The prevalence of BRCA1/2 germline varies greatly worldwide depending on various factors such as race, ethnicity etc. (Bhaskaran et al. 2019). In some populations, a wide spectrum of BRCA1/2 variants have been documented, whereas in other populations, geographically restricted variants with homogenous genomic characteristics are observed. Such homogenous genomic characteristics are expected to be observed in populations with high frequency of mating/inbreeding between genetically related individuals (Matteis et al. 2024). Therefore, in areas with high consanguinity rates/intrafamilial unions, such as in Arab countries having consanguineous marriage rates as high as 20–50% (Goundali et al. 2022), presence of geographically restricted, distinct BRCA1/2 variants, with homogenous genomic characteristics are expected. However, there are limited studies from Arab countries documenting the prevalence of BRCA1/2 variants. Furthermore, inconsistent detailing on genomic characteristics of BRCA1/2 mutational spectrum have been reported making it difficult to understand the dynamics of these variants. For example, a study from Saudi Arabia on germline BRCA variants in 61 ovarian cancer cases reported BRCA mutational positivity as 41% (Agha et al. 2022). This is a very high percent positivity from the perspective of both Saudi Arabia and international data sets. Comparing this to Jordan, a significantly different data statistics is reported. A study on 100 BC patients from Jordan reported that 27% of high-risk BC patients harboured BRCA1/2 pathogenic or likely PVs (Abdel-Razeq et al. 2018). Similarly, another study from Jordan on 192 BC and 8 ovarian cancer (OC) patient reported prevalence of pathogenic BRCA variants as 14.5% (Abu-Helalah et al. 2020). A larger study from six Middle East countries including Saudi Arabia, Lebanon, Jordan, Qatar, Egypt, and Syria reported variable prevalence between 5.6 and 20%, in high-risk BC patients indicating inconsistent prevalence rates (Abulkhair and Saghir 2021). Though, these differences might be explained as biological differences in BRCA gene prevalence, different ethnicities, differences in the age at onset in the study population etc., the main conclusion observed in majority of the studies was that a family history of breast or OC, is a significantly strong predictor/determinant of prevalent BRCA gene variants in a population (Abulkhair and Saghir 2021). Keeping in perspective the high consanguinity rates in these countries, it is expected that family history may have an impact on the prevalence of genetically distinct/novel BRCA1/2 variants in such cohorts.
In UAE, limited studies have been conducted to determine the prevalence and patient characteristics associated with BRCA variants. In one study on 27 BC patients, germline BRCA1 and BRCA2 variation rates were reported as 29.6% for BRCA1 while no germline BRCA2 variants were identified (Al-Shamsi et al. 2021). Interestingly, a study on hereditary cancer syndromes in 276 patients with various cancers from the UAE reported that 8.7 and 2.9% of patients harboured pathogenic BRCA1/2 variants. Accordingly, the study documented BRCA1/2 as the most common variation across all cancer types (accounting for 55% of the variants) indicating its role as a significant marker for understanding cancer dynamics in UAE (Dawood et al. xxxx).
In UAE, majority of studies on BRCA1/2 are mainly focused on discussing its prevalence. However, there is paucity of data on genetic characterization of BRCA1/2 variants as pathogenic/likely pathogenic (P/LP) and Variant of Uncertain Significance (VUS) in UAE population. Understanding variant categorization in these genes is critical to decipher the full spectrum of BRCA1/2 mutational associated disease dynamics. Furthermore, data on high-risk family screening cohorts, harbouring BRCA1/2 variants is missing from UAE population. Keeping in perspective the high consanguinity rates, especially in Emirati population, documentation of this data can serve as an important tool for risk management and implementation of effective screening strategies.
Our study aims to comprehensively analyse the prevalent germline BRCA1/2 variants influencing breast, ovarian, prostate, pancreatic cancer and high-risk family screening cohorts in Dubai emirate of the UAE. The results from this study aim to serve as a foundation for conducting larger studies from UAE for implementation of robust management and screening strategies.
Materials and methods
This exploratory, descriptive study included chart review of all patients that were counselled or treated in Dubai Hospital, UAE and had undergone BRCA1 and BRCA2 genetic variant analysis in the Emirate of Dubai. The study was approved by the Dubai Scientific Research Ethics Committee (DSREC) of Dubai Health Authority, UAE.
A total of 443 participants were included in the study, with study population comprising of two groups:
Group 1: Patients with confirmed diagnosis of either breast, ovarian, prostate, and pancreatic cancer (n = 306).
Group 2: healthy family members known as the family screening cohort (considered as high-risk cohort due to strong indications of family history of cancers) (n = 137).
Both patient cohort groups included Emirati nationals and non-Emirati Arab and Non-Arab expatriates (Table 1). Emiratis are defined as citizen population of the United Arab Emirates (UAE Nationals) while non-Emirati cohort are defined as all patients belonging to nationalities other than UAE. This non-Emirati cohort was further stratified as Arab and non-Arabs expats.
Table 1.
Baseline demographics of Group1 Cancer and Group 2 Family Screening Cohort
| Groups | Median Age years (range) | Age classification | Gender n (%) | Ethnicity | |||
|---|---|---|---|---|---|---|---|
| ≥ 40 years | < 40 years | Males | Females | Emiratis | Non-Emiratis | ||
| Group 1 (n = 306) Cancer Cohort | 48 (23–96) | 247 (81) | 59 (19) | 35 (11) | 271 (89) | 148 (48) | 158 (52) |
| Breast Cancer (n = 240) | 46 (29–96) | 184 (77) | 56 (23) | 5 (2) | 235 (98) | 106 (44) | 134 (56) |
| Ovarian ( n = 32) | 51 (30–76) | 30 (94) | 2 (6) | N/A | 32 (100) | 19 (59) | 13 (41) |
| Prostate (n = 19) | 71 (58–79) | 19 (100) | – | 19 (100) | N/A | 14 (74) | 5 (26) |
| Pancreatic (n = 15) | 48 (23–69) | 14 (93) | 1 (7) | 11 (73) | 4 (27) | 9 (60) | 6 (40) |
| Group 2 (n = 137) Family screening cohort | 41 (12–72) | 4 (2.9) | 133 (97.1) | 102 (74) | 35 (26) | ||
Sample preparation and next generation sequencing
Peripheral blood samples of all patients were collected at the request of the ordering physician in an EDTA anti-coagulant collection tube. Genomic DNA (gDNA) was isolated from peripheral blood leukocytes using EZ1 Advanced XL DSP DNA Blood Cards in combination with EZ1 DSP DNA Blood Kits on an automated DNA extraction system on EZ1 Advanced XL machine (Qiagen, Hilden, Germany). All extracted gDNA were quality controlled on Nanodrop One (ThermoFisher, Waltham, Massachusetts, United States) and quantified using Qubit 3.0 fluorometer using HS dsDNA assay kit (ThermoFisher, Waltham, Massachusetts, United States). A 100 ng of dsDNA from each patient was sheared on Covaris M220 Focused-ultrasonicator (Covaris LLC, Woburn, Massachusetts, United States) and the library was prepared using SureSelect Community Design Glasgow Cancer Panels—a hybrid capture-based probe (Agilent, Santa Clara, California, United States) and ran on automated Magnis NGS Prep System by Agilent using the SureSelect XTHS RevB program without modifying the preloaded configuration. The pooled libraries were quantified and sequenced as pair-end on the Illumina NextSeq500 system. FASTQ files were generated using local run manager 2.0 (Illumina, Inc., San Diego, California, United States) and aligned to hg38 reference sequence using Illumina cloud-based BaseSpace with Dragen Enrichment App v3.10.4. (Richards et al. 2015).
Variant classification and in-silico prediction tools
Genetic sequence variants are classified according to International Agency for Research on Cancer (IARC) which classifies each gene into pathogenic, likely pathogenic, variant of uncertainty (VUS), likely benign and benign. To classify BRCA1/2 variants Illumina online Variant Interpreter annotation tool was used, and variants were reported based on ACMG guideline (Richards et al. 2015). In-silico prediction tools were used to further characterize the variants. The tools used included a) Sorting Intolerant from Tolerant (SIFT) to predict whether an amino acid substitution affects protein function based on sequence homology and the physical properties of amino acids b) MutationTaster to evaluate DNA sequence variants for their disease-causing potential c) Rare Exome Variant Ensemble Learner (REVEL) for predicting pathogenicity of missense variants and d) The Genome Aggregation Database (gnomAD), a tool that aggregates and harmonizes both exome and genome data from a wide range of large-scale human sequencing projects with their potential functional impact. Variant frequencies and analysis were calculated using EasyMedStat (version 3.21.4; www.easymedstat.com) and SPSS 22.0 (IBM, USA).
Results
Baseline demographics of Group 1 and Group 2 cohort
Baseline demographics of both groups is presented in Table 1. Emiratis comprised 48% while 52% comprised of non-Emiratis (Arabs and non-Arabs). In the non-Emirati cohort, 41% were Arabs (majorly from Egypt and Sudan) while the remaining 59% belonged to non-Arab nationalities (majorly India, Pakistan, Sri Lanka and Philippines).
The median age of the patients was calculated based on the age at diagnosis. It was observed that in group 1, patients were of younger age i.e. 48 years of age. Stratification based on cancer types showed that breast, ovarian and pancreatic cancer patients were of younger age (median ages between 46 and 51 years) while those with prostate cancer were of older age (71 years). Interestingly, a large percentage of patients with BC (23%) were classified as < 40 years of age indicating BC onset at a younger age. Five cases (2%) of male BC were also observed. The observed percent positivity for BC subtypes is shown in Fig. 1.
Fig. 1.
Percent positivity of breast cancer subtypes in Group1 Cancer cohort
Results based on ethnicity showed that high number of prostate, pancreatic and ovarian cancer patients (74%, 60%, 59% respectively) were of Emirati origin while BC was more commonly observed (56%) in non-Emiratis (Arab and non-Arab patients).
Similar to Group1, group 2 family screening cohort was also younger i.e. 41 years (range 12–72 years) with majority of them being females (97.1%) of Emirati origin (74%) indicating that Emirati females are more aware and vigilant, with counselling by primary physicians leading to compliance for genetic testing.
Germline BRCA1 and BRCA2 variants in Group 1 Cancer cohort
The frequency of germline BRCA1 and BRCA2 variants observed in Group 1 Cancer cohort is shown in Table 2. A total of 38 samples (38/306; 12.4%) harboured pathogenic/likely pathogenic or VUS in this cohort. Out of 38, 23 samples (7.5%) were classified as positive for P/LP for germline BRCA1 or BRCA2 gene while the remaining 15 samples (4.9%) harboured VUS in BRCA1 or BRCA2 gene.
Table 2.
BRCA variants observed in Group 1 and Group 2 cohort
| Groups (n) | Positive (P/LP) | VUS |
|---|---|---|
| n (%) | ||
| Group 1 Cancer Cohort (306) | 23 (7.5) BRCA1: 10 (43) BRCA2: 13 (57) |
15 (4.9) BRCA1: 3 (20) BRCA2: 11 (73) BRCA1/2: 1(7) |
| Breast Cancer (240) |
17 (7.1) BRCA1: 7 (41) BRCA2:10 (59) |
13 (5.4) BRCA1: 2 (15.3) BRCA2: 10 (77) BRCA1/2: 1 (7.7) |
| Ovarian (32) |
4 (12.5) BRCA1: 3 (75) BRCA2: 1 (25) |
1 (3.1) BRCA2: 1 (100) |
| Prostate (19) |
2 (10.5) BRCA2: 2 (100) |
1 (5.3) BRCA1: 1 (100) |
| Group 2 Family Screening Cohort (137) |
5 (3.6) BRCA1: 3 (60) BRCA2: 2 (40) |
9 (6.6) BRCA1:1 (11) BRCA2: 8 (89) |
P/LP pathogenic/likely pathogenic, VUS Variants of Uncertain significance
Out of 23 P/LPs, 10 PVs were observed in BRCA1 gene (10/23; 43%) while 13 samples (13/23; 57%) were classified as P/LP for BRCA2 gene.
Similarly, out of 15 VUS, three samples (3/15; 20%) harboured VUS in BRCA1 gene, 11 samples (11/15; 73%) harboured VUS in BRCA2 gene and one sample (1/15; 7%) harboured VUS in both BRCA1/22 gene. (Fig. 2A and B). It was also observed that VUS in combination with P/LP were not detected in the same patients in both Group1 and Group 2 cohorts.
Fig. 2.
A, B Variant types observed in BRCA1 and BRCA2 gene
BRCA1 pathogenic variants in Group1 Cancer cohort
Stratified data on germline BRCA1 and BRCA2 P/LP variants is given in Table 2 and 3. In ten samples with BRCA1 PVs, majority of variations were distributed in breast (TNBC; three out of ten, ER/PR; two out of ten, PR; two out of ten) and OC (three out of ten). Out of these, four patients were of Emirati origin, two were of non-Emirati Arab origin (Egyptian and Yemeni) and four were of non-Emirati non-Arab origin (Indian, Pakistani and Philippines). All patients were young females with median age of 43 years (Table 4).
Table 4.
Pathogenic/Likely pathogenic germline BRCA1 and BRCA2 variants observed in Group1 Cancer Cohort
| Cancer Type |
Ethnicity | Age (yrs) | Gender | Family History | HGVS variant |
Chr location (hg38) |
Exon number |
Predicted Effect Protein level |
Classification | ||
|---|---|---|---|---|---|---|---|---|---|---|---|
| ACMG | ClinVar ID |
dbSNP | |||||||||
| Germline BRCA 1 pathogenic variants | |||||||||||
|
Breast TNBC |
Emirati | 44 | Female | ND | c.3228_3229delAG | chr17:43,092,302 | 10 | p.Gly1077AlafsTer8 | PVS1 + PP5 + PM2 | 37,516 | rs80357635 |
| Emirati | 54 | Female | Yes | c.4065_4068del | chr17:43,091,463 | 10 | p.Asn1355LysfsTer10 |
PVS1 + PP5 + PM1 + PM5 + PM2 |
17,674 | rs80357508 | |
| Indian | 43 | Female | Yes | c.68_69delAG | chr17: 43,124,028 | 2 | p.Glu23ValfsTer17 | PVS1 + PP5 + PM2 | 17,662 | rs80357914 | |
| Breast PR + | Egyptian | 32 | Female | ND | c.1224del | chr17:43,094,307 | 10 | p.Val409Ter | PVS1 + PP5 + PM2 | 252,436 | rs879255320 |
| Pakistani | 35 | Female | ND | c.4183C > T | chr17:41,242,963 | 11 | p.Gln1395Ter | PVS1 + PP5 + PM2 | 55,125 | rs80357260 | |
|
Breast ER + PR + |
Philippine | 35 | Female | ND | c.5251C > T | chr17:41,209,095 | 19 | p.Arg1751Ter |
PP5 + PVS1 + PS3 + PM2 |
55,480 | rs80357123 |
| Philippine | 40 | Female | ND | c.5314C > T | chr17:43,057,078 | 20 | p.Arg1772Ter | PVS1 + PP5 + PS3 + PM2 | 55,480 | rs80357123 | |
| Ovarian | Emirati | 48 | Female | Yes | c.4065_4068del | chr17:43,091,463 | 10 | p.Asn1355LysfsTer10 | PVS1 + PP5 + PM2 | 17,674 | rs80357508 |
| Emirati | 71 | Female | Yes | c.4065_4068del | chr17:43,091,463 | 10 | p.Asn1355LysfsTer10 | PVS1 + PP5 + PM2 | 17,674 | rs80357508 | |
| Yemeni | 52 | Female | ND | c.1140dup | chr17:43,094,390 | 10 | p.Lys381GlufsTer3 | PVS1 + PP5 + PM2 | 231,732 | rs876659327 | |
| Germline BRCA2 pathogenic/likely pathogenic variants | |||||||||||
|
Breast ER + PR + |
Emirati | 41 | Female | No | c.262_263del | chr13:32,319,269 | 3 | p.Leu88AlafsTer12 | PVS1 + PP5 + PM2 | 51,317 | rs276174825 |
| Emirati | 51 | Female | No | c.262_263del | chr13:32,319,269 | 3 | p.Leu88AlafsTer12 | PVS1 + PP5 + PM2 | 51,317 | rs276174825 | |
| Emirati | 33 | Female | ND | c.5645C > A | chr13:32,340,000 | 11 | p.Ser1882Ter | PVS1 + PP5 + PM2 | 37,984 | rs80358785 | |
| Iraqi | 52 | Female | ND | c.7007G > A | chr13:32,346,896 | 13 | p.Arg2336His | PP5 + PS3 + PM5 + PP3 + PM2 | 38,077 | rs28897743 | |
| Pakistani | 54 | Male | Yes | c.5222_5225del | chr13:32,339,574 | 11 | p.Ser1741ThrfsTer35 | PM2 + PMS + PVS1 | 126,063 | rs80359498 | |
| Pakistani | 42 | Female | ND | c.9435_9436del | chr13:32,394,864 | 25 | p.Ser3147CysfsTer2 | PVS1 + PP5 + PM2 | 38,240 | rs80359763 | |
| Indian | 40 | Female | ND | c.9276 T > G | chr13:32,394,708 | 25 | p.Tyr3092Ter |
PVS1 + PP5 + PM2 + PM3 |
52,803 | rs80359197 | |
|
Breast ER + PR + HER2 + |
Emirati | 58 | Female | Yes | insertion of 1.8 Kb genomic sequence of ARL13A | chr13:32,931,928/chrX:100,226,563–100228377 (hg19) | 16 | Unknown | PVS1 + PM2 | – | – |
|
Breast TNBC |
Emirati | 52 | Female | ND | c.7006C > T | chr13:32,346,895 | 13 | p.Arg2336Cys | PM2 + PMS + PVS1 | 96,845 | rs431825347 |
| Srilankan | 41 | Female | No | c.4003G > T | chr13:32,338,358 | 11 | p.Glu1335Ter | PVS1 + PM2 | 233,112 | rs747070579 | |
| Ovarian | Emirati | 47 | Female | Yes | c.8023A > G | chr13:32,363,225 | 18 | p.lle2675Val | PP5 + PS3 + PM2 + PP3 + BP1 | 52,475 | rs397507954 |
| Prostate | Emirati | 72 | Male | ND | c.7558C > T | chr13:32,356,550 | 14 | p.Arg2520Ter | PVS1 + PP5 + PM2 | 52,353 | rs80358981 |
| Palestinian | 62 | Male | ND | c.2254_2257del | chr13:32,336,606 | 11 | p.Asp752PhefsTer19 | PVS1 + PP5 + PM2 | 51,260 | rs80359326 | |
HGVS Human Genome Variation Society, hg38 Genome Reference Consortium Human Build 38, Chr Chromosome, ACMG American College of Medical Genetics and Genomics, dbSNP Single Nucleotide Polymorphism database, TNBC Triple Negative Breast Cancer, PR progesterone, ER/PR estrogen/progesterone, HER2 human epidermal growth factor receptor 2, ND Not disclosed
Majority of the Emirati patients had a positive family history of cancer while for one Emirati patient, family history of cancer could not be determined. In all Emirati patients, deletion, frameshift variants (four out of ten samples) were observed with three of them harbouring variant c.4065_4068del (p.Asn1355LysfsTer10; Exon 10). This variant was detected in TNBC (one out of three; 33%) and OC patients (two out of three; 67%) with positive family history indicating its role in these cancer types and familial genetics. One Emirati patient, diagnosed with TNBC and unknown family history of cancers, harboured a distinct variant c.3228_3229delAG (p.Gly1077AlafsTer8; Exon 10). Moreover, all the BRCA1 variants observed in Emirati patients were classified as strongly and moderately pathogenic (for e.g. PVS1 + PP5 + PM2) as per ACMG guidelines.
In non-Emirati Arab/non-Arab cohort (six out of ten), a distinct pattern of Nonsense (four out of six) and frameshift variants (two out of six) were observed. The non-sense variants were observed in various BC subtypes while frame shift variants were observed in TNBC and OC patients only.
In this cohort, distinct BRCA1 variants at various exon numbers (Shelpai 2019; Network and (NCCN). 2025; ; ; Bhaskaran et al. 2019; Dawood et al. 2025; Richards et al. 2015) were observed (Table 4) indicating that non-Emirati patients have variable BRCA1 mutational patterns that do not show similarity with the variants observed in Emirati patients. Also, in this cohort, family history was positive for one patient only (TNBC; Indian origin; frameshift variant c.68_69delAG; p.Glu23ValfsTer17; Exon 2) while for the remaining patients, family history could not be determined. All BRCA1 variants observed in non-Emirati Arab/non-Arab cohort were classified as strongly and moderately pathogenic as per ACMG guidelines.
BRCA2 pathogenic variants in Group1 Cancer cohort
Stratification of BRCA2 P/LP variants is shown in Tables 2, 3 and 4. In 13 samples with BRCA2 PVs, majority of variants were distributed in BC (TNBC; two out of thirteen, ER/PR; seven out of thirteen, ER/PR/HER2; one out of thirteen), ovarian (one out of thirteen) and prostate cancer patients (two out of thirteen). Similar to BRCA1, majority of BRCA2 variants were observed in BC cohort.
Table 3.
BRCA variants in Breast cancer subtypes in Group 1 cancer cohort
| Breast cancer subtypes (n = 30) | Positive (P/LP) | VUS |
|---|---|---|
| n (%) | ||
| ER + PR + (n = 14) |
9 (64.3) BRCA1: 2 (22) BRCA2: 7 (78) |
5 (36) BRCA1: 1 (20) BRCA2: 4 (80) |
|
TNBC (n = 7) |
5 (71.4) BRCA1: 3 (60) BRCA2: 2 (40) |
2 (29) BRCA2: 1(50) BRCA1/2: 1(50) |
|
PR + (n = 2) |
2 (100) All BRCA1 |
- |
|
ER + PR + HER2 + (Triple positive) (n = 1) |
1 (100) All BRCA2 |
- |
|
HER2 + (n = 1) |
- |
1 (100) All BRCA1 |
|
ER + (n = 4) |
- |
4 (100) All BRCA2 |
|
ER + HER2 + (n = 1) |
- |
1 (100) All BRCA2 |
P/LP pathogenic/likely pathogenic, VUS Variants of Uncertain significance, TNBC Triple Negative Breast Cancer, PR progesterone, ER/PR estrogen/progesterone, HER2 human epidermal growth factor receptor 2
Out of 13 BRCA2 P/LP variant harbouring patients, seven were of Emirati origin, two were of non-Emirati Arab origin (Iraqi and Palestinian) and four were of non-Emirati non-Arab origin (Indian, Pakistani and Sri Lankan). All the female patients (breast and ovarian cancer) were of younger age with median age of 49 years. With respect to prostate cancer, the age distribution was higher (mean age 67 years). Varied family history was observed with two out of seven Emirati patients having a positive family history of cancer, two having negative family history, while for remaining three patients, family history could not be determined. In non-Emirati Arab/non-Arab cohort of six patients, only one patient had a positive family history while for the remaining, family history could not be determined (Table 4).
In seven Emirati patients, variable mutational patterns were observed. In two out of seven patients with ER + PR + BC, deletion frameshift variant c.262_263del (p.Leu88AlafsTer12; Exon 2) was observed. Family history was negative for these patients indicating other factors at play. Missense variation was observed for two out of seven patients (one TNBC and one OC; exon 13 and 18). One Missense variant (c.7006C > T; p.Arg2336Cys) observed in TNBC patient was classified as likely pathogenic by ACMG classification although in-silico prediction showed conflicting result for this variant (SIFT tolerated and PolyPhen benign). However, for the other distinct missense variants (c.8023A > G; p.lle2675Val) observed in OC patient with family history of cancer, in-silico prediction classified it as SIFT damaging and MutationTaster damaging. This variant was classified as pathogenic by ACMG classification as well. Nonsense variants were observed in two out of seven Emirati patients (Prostate and ER + PR + BC; exon 11 and 14). Interestingly, in one ER + PR + HER2 + (Triple positive) BC patient, with a positive family history, a large insertion of 1.8 Kb genomic sequence of ARL13A at chromosome13 and chromosome X location 100,226,563–100228377 (hg19) Exon 16 was observed. This is a distinct variant that could be linked to a familial pattern (Table 4).
Out of six BRCA2 PVs in non-Emirati Arab/non-Arab cohort, deletion frameshift variants (two out of six) and Non-sense variants (two out of six) were commonly observed in patients of non-Emirati non-Arab origin (Deletion frameshift: two Pakistani patients with ER + PR + BC; Non-sense: one Indian with ER + PR + BC, one Sri Lankan with TNBC). One of Pakistani patient harbouring deletion variant c.5222_5225del (p.Ser1741ThrfsTer35; Exon 11) was a male BC patient of young age (54 years) with a positive family history indicating a familial pattern. In non-Emirati Arab origin patients, Missense variant c.7007G > A (p.Arg2336His; Exon 13; Iraqi) was observed in ER + PR + BC with in-silico prediction classifying it as SIFT damaging and MutationTaster disease causing. This variant was classified as pathogenic by ACMG guidelines. The deletion frameshift variant c.2254_2257del (p.Asp752PhefsTer19; Exon 11) was observed in a Palestinian male with prostate cancer.
BRCA1 variants of uncertain significance in Group1 cancer cohort
Stratified data on germline BRCA1 and BRCA2 VUS is given in Tables 2, 3 and 5. All BRCA1 VUS were observed in Emirati patients. Among these, two patients had a positive family history of cancer. For all patients, missense variant was the common variation type. However, in-silico prediction with various software determined interesting results. One ER + PR + BC patient, with positive family history) harbouring variant c.1250A > G (p.Asn417Ser; Exon 10) was predicted as pathogenic (0.9154) by REVEL. Moreover, in HER2 + BC patient (negative family history) harbouring VUS c.5260G > C (p.Glu1754Gln; Exon 20), in-silico SIFT predicted it as pathogenic (0.9125) and disease causing (0.81). Further functional studies can shed more light on these variants and their significance in Emirati BC cohorts. Similarly, prostate cancer patient (positive family history), harbouring VUS c.1771A > G (p.Ile591Val; Exon 10) was predicted by in-silico software as PolyPhen possibly damaging (0.456) indicating that amino acid substitution is likely to have damaging effect on protein function (Table 5). Though observed in a small set of patients, these findings provide important insights into the role of BRCA1 VUS in the group1cancer cohort. Larger studies with Emirati patients from various cancer cohorts, can provide broader insight on these VUS. On the other hand, functional studies can help to determine the true significance of these in-silico identified pathogenic/damaging/deleterious VUS.
Table 5.
germline BRCA1/2 Variants of Uncertain Significance in Group1 Cancer Cohort
| Cancer Type |
Ethnicity | Age (yrs) | Gender | Family History | HGVS variant |
Chr location (hg38) |
Exon number |
Predicted Effect Protein level |
Classification | ||
|---|---|---|---|---|---|---|---|---|---|---|---|
| ACMG | ClinVar ID |
dbSNP | |||||||||
| Germline BRCA 1 VUS | |||||||||||
|
Breast ER + PR + |
Emirati | 51 | Female | Yes | c.1250A > G | chr17:43,094,281 | 10 | p.Asn417Ser | PM2 + BP4 | 54,174 | rs80357113 |
|
Breast HER2 + |
Emirati | 41 | Female | No | c.5260G > C | chr17:43,057,069 | 20 | p.Glu1754Gln | PM2 + PM1 | 462,666 | rs80357432 |
| Prostate | Emirati | 75 | Male | Yes | c.1771A > G | chr17:43,093,760 | 10 | p.Ile591Val | PM2 | 421,781 | rs1064795358 |
| Germline BRCA2 VUS | |||||||||||
|
Breast ER + |
Emirati | 63 | Female | No | c.9104A > C | chr13:32,379,900 | 23 | p.Tyr3035Ser | PM2 + PP3 | 38,211 | rs80359165 |
| Emirati | 66 | Female | No | c.6842G > A | chr13:32,344,558 | 12 | p.Gly2281Glu | PM2 + PP3 | 926,077 | rs80358908 | |
| Algerian | 46 | Female | No | c.8782G > T | chr13:32,379,344 | 22 | p.Ala2928Ser | PM2 + PP3 | 141,459 | rs587781762 | |
| Philippines | 42 | Female | No | c.1826A > G | chr13:32,333,304 | 10 | p.Gln609Arg | PM2 | 51,210 | rs80358473 | |
|
Breast ER + PR + |
Emirati | 57 | Female | No | c.5474C > T | chr13:32,339,829 | 11 | p.Ala1825Val | PM2 + BP4 | 37,968 | rs397507352 |
| Emirati | 50 | Female | No | c.3762G > T | chr13:32,338,117 | 11 | p.Glu1254Asp | PM2 + BP4 | 231,869 | rs777028631 | |
| Iranian | 43 | Female | No | c.7628A > G | chr13:32,357,752/ | 16 | p.Tyr2543Cys | PM2 + BP1 | 102,758 | rs431825354 | |
| Indian | 34 | Female | Yes | c.8117A > G | chr13:32,363,319 | 18 | p.Asn2706Ser | PM2 + BP4 | 38,139 | rs80359055 | |
|
Breast TNBC |
Emirati | 50 | Female | No | c.8117A > G | chr13:32,363,319 | 18 | p.Asn2706Ser | PM2 + BP4 | 38,139 | rs80359055 |
|
Breast ER + HER2 + |
Uganda | 43 | Female | No |
c.7676C > G c.7712A > G |
chr13:32,357,800 | 16 |
p.Ser2559Cys p.Glu2571Gly |
PM2 + PP3 |
409,479 52,392 |
rs1060502421 rs55689095 |
| Ovarian | Sudani | 56 | Female | No | c.4574A > T | chr13:32,338,929 | 11 | p.His1525Leu | PM2 + BP4 | 838,870 | rs397507336 |
| Germline BRCA1/2 VUS | |||||||||||
| Breast TNBC | Indian | 31 | Female | No |
BRCA1: c.1593G > A BRCA2: c.147A > C |
chr17:43,093,938 chr13:32,319,156 |
10 3 |
p. Met531Ile p. Glu49Asp |
PM2 + BP4 |
3,074,124 920,387 |
rs779648876 |
HGVS Human Genome Variation Society, hg38 Genome Reference Consortium Human Build 38, Chr Chromosome, ACMG American College of Medical Genetics and Genomics, dbSNP Single Nucleotide Polymorphism database, TNBC Triple Negative Breast Cancer, PR progesterone, ER/PR estrogen/progesterone, HER2 human epidermal growth factor receptor 2, ND Not disclosed
BRCA2 variants of uncertain significance in gene in Group1 cancer cohort
In BRCA2, 11 samples harboured VUS, with majority of VUS observed in BC patients (Tables 2, 3 and 5). Majority of the patients had no family history of cancer (Table 5).
In two Emirati ER positive BC patients, VUS c.9104A > C (p.Tyr3035Ser; Exon 23) and c.6842G > A (p.Gly2281Glu; Exon 12), SIFT predicted both these as PolyPhen: probably damaging indicating deleterious effect of these variants on protein function. In non-Emirati Arab origin (Algerian and Sudani origin) cohort, similar results were observed for ER + breast and OC patients. The VUS variants c.8782G > T (p.Ala2928Ser; Exon 22) and c.4574A > T (p.His1525Leu; Exon 11) were classified as PolyPhen: probably damaging with predicted deleterious effects on the protein function. Interestingly, in a young patient, 43-year-old with no family history of cancer, with ER + HER2 + BC, two VUS variants at Exon 16 were observed in BRCA2 gene. Both the identified variants c.7676C > G (p.Ser2559Cys) and c.7712A > G (p.Glu2571Gly) were predicted as PolyPhen: probably damaging. Of these c.7712A > G (p.Glu2571Gly) was predicted as having deleterious effects on protein function. Interestingly, for this variant, REVEL predicted it as pathogenic as well indicating this VUS can be of significance for further studies with larger sample size/ functional analysis to help understand its role in BRCA2 associated BC pathogenesis. All the other VUS were predicted as benign through in-silico prediction. In one of the patients, VUS were observed in both BRCA1 and BRCA2 genes, but both were classified as benign via in-silico prediction (Table 5).
Germline BRCA1 and BRCA2 variants in Group 2 family screening cohort
The frequency of germline BRCA1 and BRCA2 variants observed in Group 2 family screening cohort is shown in Tables 2 and 6. A total of 14 samples (14/137; 10%) harboured BRCA1 and BRCA2 pathogenic variants or VUS in the respective group. Out of 137 samples, five samples (3.6%) were found to be pathogenic for germline BRCA1 or BRCA2 gene while the remaining nine samples (6.6%) harboured VUS in BRCA1 or BRCA2 gene.
Table 6.
germline BRCA1/2 pathogenic variants and VUS in Group 2 Family screening Cohort
| Ethnicity | Age (yrs) | Gender | Family History | HGVS variant |
Chr location (hg38) |
Exon number |
Predicted Effect Protein level |
Classification | ||
|---|---|---|---|---|---|---|---|---|---|---|
| ACMG | ClinVar ID |
dbSNP | ||||||||
| Germline BRCA1 pathogenic variants | ||||||||||
| Emirati | 44 | Female | Yes | c.3756_3759del | chr17:43,091,772 | 10 | p.Ser1253ArgfsTer10 | PVS1 + PM2 + PP5 | 17,673 | rs80357868 |
| Emirati | 49 | Female | Yes | c.4065_4068del | chr17:43,091,463 | 10 | p.Asn1355LysfsTer10 | PVS1 + PP5 + PM2 | 17,674 | rs80357508 |
| Sudani | 65 | Male | Yes | c.5095C > T | chr17:43,063,931 | 17 | p.Arg1699Trp |
PS3 + PM1 + PP5 + PM5 + PP3 + PM2 |
55,396 | rs55770810 |
| Germline BRCA2 pathogenic variants | ||||||||||
| Emirati | 45 | Female | Yes | c.771_775del | chr13:32,331,004 | 9 | p.Asn257LysfsTer17 | PVS1 + PP5 + PM2 | 9326 | rs80359671 |
| Bahraini | 42 | Female | Yes | c.8377G > A | chr13:32,370,447 | 19 | p.Gly2793Arg |
PP5 + PP3 + PM2 + BP1 |
52,569 | rs80359082 |
| Germline BRCA1 VUS | ||||||||||
| Emirati | 47 | Female | Yes | c.1396C > G | chr17:43,094,135 | 10 | p.Aly466Gly | PM2 | 54,240 | rs80356964 |
| Germline BRCA2 VUS | ||||||||||
| Emirati | 17 | Female | Yes | c.2657A > G | chr13:32,337,012 | 11 | p.Asn886Ser | PM2 + BP4 | 91,782 | rs80358526 |
| Emirati | 21 | Female | Yes | c.2657A > G | chr13:32,337,012 | 11 | p.Asn886Ser | PM2 + BP4 | 91,782 | rs80358526 |
| Emirati | 24 | Female | Yes | c.2657A > G | chr13:32,337,012 | 11 | p.Asn886Ser | PM2 + BP4 | 91,782 | rs80358526 |
| Emirati | 25 | Female | Yes | c.2657A > G | chr13:32,337,012 | 11 | p.Asn886Ser | PM2 + BP4 | 91,782 | rs80358526 |
| Emirati | 29 | Female | Yes | c.2892A > T | chr13:32,337,247 | 11 | p.Lys964Asn |
Likely Benign BP4 + BP1 + PM2 |
133,728 | rs587778119 |
| Emirati | 37 | Female | Yes | c.2892A > T | chr13:32,337,247 | 11 | p.Lys964Asn | PM2 + BP4 | 133,728 | rs587778119 |
| Emirati | 44 | Female | Yes | c.2892A > T | chr13:32,337,247 | 11 | p.Lys964Asn | PM2 + BP4 | 133,728 | rs587778119 |
| Jordanian | 47 | Female | Yes | c.1472C > G | chr13:32,332,950 | 10 | p.Thr491Ser | PM2 + BP4 | 37,742 | rs397507268 |
HGVS Human Genome Variation Society, hg38 Genome Reference Consortium Human Build 38, Chr Chromosome, ACMG American College of Medical Genetics and Genomics, dbSNP Single Nucleotide Polymorphism database
Out of five PVs in BRCA1 and BRCA2 genes, three PVs were observed in BRCA1 gene (60%) while two samples (40%) were classified as pathogenic for BRCA2 gene. Similarly, out of nine samples classified as VUS, one sample (11%) harboured VUS in BRCA1 gene while eight samples (89%) harboured VUS in BRCA2 gene. All patients had a positive family history of cancers.
BRCA1 pathogenic variants in Group 2 family screening cohort
Stratified data on germline BRCA1 PVs is given in Table 6. In three samples with BRCA1 pathogenic variations, majority of variations (two out of three) were observed in young Emirati individuals (44 and 49 years old) with common variant type observed as deletion variant. Moreover, in one older age (65 years old) non-Emirati Arab origin (Sudani) male individual, pathogenic missense variant c.5095C > T (p.Arg1699Trp; Exon 17) was observed. In-silico SIFT predicted this variant as deleterious and PolyPhen: possibly damaging (0.885) indicating the significance of this variant in this high-risk screening. Further studies on family screening cohorts can give a broad perspective on this variant for management purposes.
BRCA2 pathogenic variants in Group 2 family screening cohort
In Emirati patient, BRCA2 pathogenic deletion variant c.771_775del (p.Asn257LysfsTer17; Exon 9) was observed in young (45 year old) Emirati individual. However, interesting results were observed in a young (42 years old) non-Emirati Arab (Bahraini origin) individual with missense variant c.8377G > A (p.Gly2793Arg; Exon 19). In this patient, in-silico SIFT predicted it as deleterious (0.9125), PolyPhen: Damaging (1.00) and pathogenic via REVEL indicting the significance of this variant in predisposition to cancers (Table 6).
BRCA1 variants of uncertain significance in Group 2 family screening cohort
Stratified data on germline BRCA1 and BRCA2 VUS is given in Tables 2 and 6. In BRCA1, only one sample from young (47 years old) Emirati individual was positive for missense variant c.1396C > G (p.Aly466Gly; Exon 10). This variant was predicted as PolyPhen: possibly damaging (0.355) (Table 6).
BRCA2 variants of uncertain significance in Group 2 family screening cohort
Out of eight patients harbouring VUS in BRCA2 gene, majority (even out of eight; 87.5%) were young Emirati females, with a median age of 25 years. Common VUS (four out of seven; 57%) observed was missense c.2657A > G (p.Asn886Ser; Exon 11) variant with in-silico prediction of PolyPhen: probably damaging (0.985). In addition to this, another VUS, c.2892A > T (p.Lys964Asn; Exon 11) was commonly observed in three Emirati females (43%) with in-silico analysis predicting it as PolyPhen: possibly damaging (0.355) for two of these individuals. Identification of common VUS in BRCA2 indicate in Emirati family screening cohort indicate their importance in this cohort and could be investigated further as part of a larger screening with functional studies.
Impact of body mass index (BMI) on cancer risk
In our study, we did not observe any significant difference in the distribution of BMI between cases and controls (28.7 vs. 27.9 kg/m2, p value 0.180). Furthermore, a similar trend was observed upon stratified analysis by cancer subtype (BC: 28.9 vs. 27.9 kg/m2, P = 0.140; OC: 29.1 vs. 27.9 kg/m2, p-value 0.394, pancreatic cancer: 28.3 vs. 27.9 kg/m2, p value 0.805; prostate cancer: 27.6 vs. 27.9 kg/m2, p value 0.852). Therefore, no association was observed between BMI and cancer risk among BRCA1/2 carriers (p value 0.688).
Discussion
Our study aimed to explore and report on the germline BRCA1 and BRCA2 variants in cancer and family screening cohort in the Emirates of Dubai, UAE. We observed a range of P/LP variants and VUS in both Emirati and non-Emirati Arab/non-Arab cohorts in both cancer and high-risk family screening cohorts. The findings provide insights into the prevalence and distribution of BRCA1 BRCA2 variants in an ethnically diverse population residing in the UAE. o our knowledge, comprehensive data on BRCA1/2 variants in various cancer types and family screening cohorts have not previously been published from the UAE, making this study novel.
In our cancer cohort (Group 1), the majority of patients were diagnosed with BC, primarily at a younger age (median age: 48 years). This observation is consistent with other published data from the UAE. For example, a study from Dubai reported a younger median age (44 years) in their BC cohort (Dawood et al. xxxx). Similarly, another study from Al-Ain (another Emirates in UAE) reported a BC cohort with median age 42 years, indicating a younger age onset of BC in UAE (Altinoz et al. 2020). A recommendation paper published from UAE emphasized that BC screening in UAE should commence at 30 years of age instead of 40 years, as UAE population is diagnosed with BC a decade earlier compared to Western population (Al-Shamsi 2018). Keeping this in perspective, large-scale studies for screening of high-risk families and cancer cohorts (especially with family history of cancers) from all seven emirates of UAE becomes imperative to gather evidence-based data on age related BC dynamics. This will allow implementation of targeted health awareness, screening and management strategies for younger age cohorts.
With respect to BC subtypes ER + /PR + , HER2 + and TNBC were most frequently observed in patients (69%, 23% and 18% respectively). Interestingly, the percent positivity for TNBC is towards a higher range i.e.18% in our study cohort. The global range is between 15–20% and TNBC trends in the study cohort may be driven by ethnic and regional background. In Emirati patients, the major BC subtype observed was ER + /PR + (77%) followed by HER2 + (14%) and TNBC (6%). Interestingly, in non-Emirati patients, a distinct pattern of high TNBC frequency (13%) was observed.
We observed a higher frequency of BRCA2 variants compared to BRCA1, which is consistent with other studies from the UAE and other Arab countries (Altinoz et al. 2020; Z S. 2020; Abdulrashid et al. 2019). Stratification of BRCA1 and BRCA2 variants according to cancer types showed that majority of BRCA1 and BRCA2 variants were associated with BC (30/240; 12.5%) with 17/30 cases (56.6%) classified as P/LP and 13/30 cases (43.3%) classified as VUS variants. Comparison with other studies from Dubai and other Emirates highlights some variation. One study from Dubai reported 21/276 (7.6%) BRCA1/2 P/LP variants (Dawood et al. 2025) while another study reported 21/396 (5.3%) cases to be positive for BRCA1/2 variant. On the other hand, two studies from Al-Ain reported starkly conflicting data. One study reported 54/130 (41.5%) positive cases for BRCA1/2 variants with 31/54 (57%) classified as P/LP and 23/54 (42.5%) cases classified as VUS (Altinoz et al. 2020) while the other study reported a significantly lower positivity with only 6/49 (12.2%) patients testing positive for BRCA1/2 variants (Ameri et al. 2024). A study from Ras-Al-Khaimah reported 38/262 (14.5%) positive cases for gBRCA PVs (Z S. 2020). Though this difference in BRCA1/2 variant positivity in various emirates of UAE can be related to sample size difference, study design/enrolment differences/detection methodologies, the data does provide evidence on varied prevalence of BRCA1/2 pathogenic and VUS variants in UAE. Thus, larger, more standardized studies across the seven Emirates are needed to accurately determine the prevalence and distribution of gBRCA1/2 variants. Such large studies can more accurately define prevalence and mutational pattern of gBRCA1/2 variants and would help align health awareness programs, screening tools and management protocols across all Emirates in UAE.
Stratification of BRCA1/2 variants according to BC subtypes showed interesting results. Majority of BRCA1/2 PLP and VUS were observed in ER + /PR + subtypes (64.3% and 36% respectively) followed by TNBC (71.4% and 29% respectively). BRCA2 P/LP variants were more common in ER + /PR + samples (78%) while in TNBC, BRCA1 P/LPs were observed in higher percentage (60%). In ER + PR + and TNBC, VUS were more commonly associated with BRCA2gene (80 and 50% respectively). Interestingly, one TNBC sample harboured VUS in both BRCA1/2genes. The role of VUS with respect to in-silico prediction will be discussed later. Comparing with other studies from UAE, similar results were observed with BRCA1 P/LP variants more commonly reported in TNBC subtype while BRCA2 PLP variants are more commonly reported in ER + /PR + subtype (Altinoz et al. 2020; Z S. 2020).
In breast, ovarian and prostate cancer, the most common P/LP variants observed were deletion leading to frameshift and non-sense. A deletion variant c.4065_4068del (p.Asn1355LysfsTer10) in BRCA1 was observed in one BC (TNBC) and two OC patients of Emirati ethnicity with a positive family history, and has been reported in previous studies from the UAE and other regions (Z S. 2020; Rahman and Zayed 2018; Bu et al. 2016; Younes and Zayed 2019; Laitman et al. 2019). Other frequently observed BRCA1 variants in breast and OC cohort included deletion variant c.68_69delAG (p.Glu23ValfsTer17), c.3228_3229delAG (p.Gly1077AlafsTer8), insertion c.1140dup (p.Lys381GlufsTer3) and nonsense variant c.5251C > T (p.Arg1751Ter). These BRCA1 variants have been previously reported and are considered as common recurring variants in Middle East and North Africa (MENA) region and European countries (Z S. 2020; Laitman et al. 2019). The other BRCA1 variants observed in our cohort are considered distinct and have not been reported elsewhere.
In BRCA2, common PV was deletion variant, c.262_263del (p.Leu88AlafsTer12) observed in Emirati patients (ER + /PR + BC) with no family history of cancer. As per our knowledge, this variant has not been reported from other emirates of UAE or from GCC countries indicating its novelty. However, other variants such as nonsense c.5645C > A (p.Ser1882Ter), missense variant c.7007G > A (p.Arg2336His) in ER + /PR + BC cohort (Emirati and Iraqi ethnicity respectively) and deletion variant c.2254_2257del (p.Asp752PhefsTer19) observed in Palestinian male with prostate cancer have been reported from GCC region and European countries (Laitman et al. 2019) validating that these are recurring BRCA2 variants. Other BRCA2 variants observed in our cohort can be considered novel for UAE as these have not been reported from other regions yet. However, larger studies would provide a more robust data to understand the prevalence of these novel variant types in UAE cohort.
In our high-risk family screening cohort, we were able to observe deletion and missense variants in BRCA1/2 genes respectively. In this cohort, pathogenic BRCA1 deletion variant c.4065_4068del (p.Asn1355LysfsTer10) was observed in an Emirati individual with family history of cancer. Comparatively, this variant was commonly observed in our cancer cohort (Emirati patients with breast and ovarian; family history of cancers) as well as from other studies indicating the prevalence of this variant (Z S. 2020; Rahman and Zayed 2018; Laitman et al. 2019). The presence of this common PV, specifically in an Emirati individual, of younger age and positive family history of cancer, indicates a strong genetic predisposition to cancers in the respective carrier. Other BRCA1 deletion variants, c.3756_3759del (p.Ser1253ArgfsTer10) and c.5095C > T (p.Arg1699Trp), though not observed in our cancer cohort, have been reported in studies from Greece, Italy, Morrocco and Saudi Arabia (Laitman et al. 2019). Similar to BRCA1, distinct BRCA2 PVs c.771_775del (p.Asn257Lysfs) was observed (Emirati patient with positive family history). Again, this variant, though not reported in our cancer cohort, has been reported in a study from Ras-Al-Khaimah emirate of UAE in high frequency (3/20 patients; 15%). In this study, the variation was observed in three members of one family (Emiratis; 2 sisters and aunty), age ≤ 45, all diagnosed with BC (Z S. 2020). On the other hand, this variant has been commonly reported in studies from Egypt and Jordon indicating the significance of this variant in cancer predisposition. A novel variant c.8377G > A (p.Gly2793Arg) observed in a Bahraini patient has not been reported as yet. In-silico prediction of this variant indicated it to be a deleterious, damaging and pathogenic indicating its role in cancer predisposition. Such findings described here, provides a strong rationale for conducting large family screening from all seven Emirates of UAE, to accurately analyse prevalence of BRCA1/2 variants in UAE. Such studies would help gear screening program and health awareness campaigns for personalized management of both Emirati and non-Emirati populations residing in UAE.
Our study also aimed to report on BRCA1/2 VUS in both cancer and high-risk family screening cohorts. Though interpretation of VUS is challenging for clinicians, due to lack of management and counselling guidelines (NCPGI 2025), reporting becomes significant to understand their frequency in a specific population. On the other hand, if certain VUS is observed in large frequency, in a geographically distinct population, with robust in-silico prediction classifying it as pathogenic, deleterious and damaging, it can serve as evidence to gear functional experimental efforts and studies to clarify their clinical significance. In our cancer cohort, missense VUS were observed in both BRCA1/2. In BRCA1, three distinct VUS were observed in patients of Emirati ethnicity with two of them having a family history of cancer. In-silico software predicted one of this VUS, c.5260G > C (p.Glu1754Gln) as pathogenic (0.9125) by Sorting Intolerant From Tolerant (SIFT) and disease-causing by MutationTaster (0.81). However, the meta predictor REVEL, predicted this VUS as uncertain (0.8055).
In BRCA2, three VUS were observed: c.9104A > C (p.Tyr3035Ser), c.6842G > A (p.Gly2281Glu), c.8782G > T (p.Ala2928Ser) were predicted as deleterious, PolyPhen and SIFT in-silico analysis prediction as probably-damaging whereas REVEL prediction as uncertain/benign. Interestingly, one Ugandan patient with ER + /HER2 + BC harboured two VUS of which one VUS c.7676C > G (p.Ser2559Cys) was predicted by SIFT as tolerated (0.5125) and PolyPhen as probably-damaging; while the other VUS c.7712A > G (p.Glu2571Gly) was predicted by SIFT as deleterious (0.7849), by PolyPhen as probably-damaging and by REVEL as Pathogenic. (0.9154). The presence of these two VUS in a patient with distinct predictions of pathogenicity and probably damaging amino acid substitutions require further investigation to understand its role in BC pathology.
The high-risk family screening cohort had majority of harboured missense VUS with in-silico prediction by SIFT as tolerated and by PolyPhen as probably damaging in both BRCA1/2 genes. In Emirati individuals, distinct BRCA1/2 VUS were observed in high frequency indicating the importance of establishing functional roles of these VUS in Emirati cohort.
Conclusion
Our study provides a comprehensive analysis of BRCA1/2 P/LP variants and VUS in Emirati and non-Emirati Arab/non-Arab cancer and high-risk family screening cohorts from the Emirates of Dubai in UAE. This novel and a foundational data underscore the need for larger studies across all seven Emirates in UAE, to facilitate management and screening strategies by local health regulatory bodies.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
We would like to acknowledge Al Zahrawi Medical Supplies LLC for support in covering the processing charge for this article.
Author contributions
AA and MA helped with data curation and analysis and with Figs. 2A and 2B. DE, SE and ML recruited patients and ML applied for ethics approval as lead PI. IM critically reviewed the manuscript. HY and AR wrote the manuscript and performed all data analysis.
Funding
Not Applicable.
Data availability
Raw sequencing data for this study were generated at the Dubai Genetics Center, Dubai Health, and are not publicly accessible to ensure compliance with patient confidentiality and privacy regulations. However, the derived data used to support the findings of this study have been included as Supplementary Data file. Data is provided within the manuscript or supplementary information files.
Declarations
Conflict of interest
The authors declare no competing interests.
Ethics approval and consent to participate
Ethical approval for this study was obtained from the Dubai Scientific Research Ethics Committee (DSREC), Dubai Health Authority, under reference number DSREC-07/2022_14. This research was conducted in full adherence to the principles outlined in the Declaration of Helsinki. All participants provided informed consent, including consent for genetic research and publication purposes.
Consent for publication
Not Applicable.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Contributor Information
Muhammad Farooq Latif, Email: Muflatif@dha.gov.ae.
Hemad Yasaei, Email: hemad@nrl.ae.
References
- Abdel-Razeq H, Al-Omari A, Zahran F, Arun B (2018) Germline BRCA1/BRCA2 mutations among high risk breast cancer patients in Jordan. BMC Cancer 18(1):152 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Abdulrashid K, AlHussaini N, Ahmed W, Thalib L (2019) Prevalence of BRCA mutations among hereditary breast and/or ovarian cancer patients in Arab countries: systematic review and meta-analysis. BMC Cancer 19(1):256 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Abu-Helalah M, Azab B, Mubaidin R, Ali D, Jafar H, Alshraideh H et al (2020) BRCA1 and BRCA2 genes mutations among high risk breast cancer patients in Jordan. Sci Rep 10(1):17573 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Abulkhair O, El Saghir NS (2021) BReast CAncer gene (BRCA): snapshot of the Middle East. Chin Clin Oncol 10(5):51 [DOI] [PubMed] [Google Scholar]
- Agha N, Alshamsan B, Al-Farsi S, Ateya HA, Almugbel FA, Alotaibi HA et al (2022) Assessing frequency and clinical outcomes of BRCA mutated ovarian cancer in Saudi women. BMC Cancer 22(1):18 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Al-Shamsi HOAS (2018) Breast cancer screening in the United Arab Emirates: is it time to call for a screening at an earlier age? J Cancer Prev Curr Res 9(3):123–126 [Google Scholar]
- Al-Shamsi HO, Abu-Gheida I, Abdulsamad AS, AlAwadhi A, Alrawi S, Musallam KM et al (2021) Molecular spectra and frequency patterns of somatic mutations in Arab women with breast cancer. Oncologist 26(11):e2086–e2089 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Al-Shamsi HO, Abdelwahed N, Al-Awadhi A, Albashir M, Abyad AM, Rafii S et al (2023) Breast cancer in the United Arab Emirates. JCO Glob Oncol 9:e2200247 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Altinoz A, Al Ameri M, Qureshi W, Boush N, Nair SC, Abdel-Aziz A (2020) Clinicopathological characteristics of gene-positive breast cancer in the United Arab Emirates. Breast 53:119–124 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ameri MA, Shanbhag NM, Bin Sumaida A, Ansari J, Trad DA, Dawoud EA et al (2024) Oncotype DX in Breast Cancer Management: Insights and Outcomes From the United Arab Emirates. Cureus 16(3):e56535 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Bhaskaran SP, Chandratre K, Gupta H, Zhang L, Wang X, Cui J et al (2019) Germline variation in BRCA1/2 is highly ethnic-specific: evidence from over 30,000 Chinese hereditary breast and ovarian cancer patients. Int J Cancer 145(4):962–973 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Bu R, Siraj AK, Al-Obaisi KA, Beg S, Al Hazmi M, Ajarim D et al (2016) Identification of novel BRCA founder mutations in Middle Eastern breast cancer patients using capture and Sanger sequencing analysis. Int J Cancer 139(5):1091–1097 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Dawood SS, Apessos A, Elkhoury M, Khan F, Khatib F, Bello M, et al. Analysis of hereditary cancer syndromes in patients from the United Arab Emirates. Journal of Clinical Oncology. 2018;36(15_suppl):e13509-e.
- De Matteis E, Tumolo MR, Tarantino P, Ciccarese M, Grassi T, Bagordo F et al (2024) Prevalence and spectrum of germline BRCA1 and BRCA2 in a cohort of ovarian cancer patients from the Salento peninsula (Southern Italy): a matter of preventive health. Oncotarget 15:134–141 [DOI] [PMC free article] [PubMed] [Google Scholar]
- El Goundali K, Chebabe M, Zahra Laamiri F, Hilali A (2022) The determinants of consanguineous marriages among the arab population: a systematic review. Iran J Public Health 51(2):253–265 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Farmer H, McCabe N, Lord CJ, Tutt AN, Johnson DA, Richardson TB et al (2005) Targeting the DNA repair defect in BRCA mutant cells as a therapeutic strategy. Nature 434(7035):917–921 [DOI] [PubMed] [Google Scholar]
- Imyanitov EN, Kuligina ES, Sokolenko AP, Suspitsin EN, Yanus GA, Iyevleva AG et al (2023) Hereditary cancer syndromes. World J Clin Oncol 14(2):40–68 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Laitman Y, Friebel TM, Yannoukakos D, Fostira F, Konstantopoulou I, Figlioli G et al (2019) The spectrum of BRCA1 and BRCA2 pathogenic sequence variants in Middle Eastern, North African, and South European countries. Hum Mutat 40(11):e1–e23 [DOI] [PubMed] [Google Scholar]
- Luo L, Keyomarsi K (2022) PARP inhibitors as single agents and in combination therapy: the most promising treatment strategies in clinical trials for BRCA-mutant ovarian and triple-negative breast cancers. Expert Opin Investig Drugs 31(6):607–631 [DOI] [PMC free article] [PubMed] [Google Scholar]
- National Comprehensive Cancer Network (NCCN) https://www.nccn.org/guidelines/nccn-guidelines.
- Oh M, Alkhushaym N, Fallatah S, Althagafi A, Aljadeed R, Alsowaida Y et al (2019) The association of BRCA1 and BRCA2 mutations with prostate cancer risk, frequency, and mortality: a meta-analysis. Prostate 79(8):880–895 [DOI] [PubMed] [Google Scholar]
- ONCOLOGY NCPGI. Genetic/Familial High-Risk Assessment: Breast, Ovarian, and Pancreatic, Version 2.2021. [DOI] [PubMed]
- Rahman S, Zayed H (2018) Breast cancer in the GCC countries: A focus on BRCA1/2 and non-BRCA1/2 genes. Gene 668:73–76 [DOI] [PubMed] [Google Scholar]
- Richards S, Aziz N, Bale S, Bick D, Das S, Gastier-Foster J et al (2015) Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American college of medical genetics and genomics and the association for molecular pathology. Genet Med 17(5):405–424 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Shelpai W. Cancer incidence in United Arab Emirates annual report of the UAE—National cancer registry MOHAP: MOHAP; 2019 www.mohap.gov.ae.
- Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A et al (2021) Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 71(3):209–249 [DOI] [PubMed] [Google Scholar]
- WHO IAfRoC. 2024 https://gco.iarc.who.int/media/globocan/factsheets/populations/784-united-arab-emirates-fact-sheet.pdf.
- Wong W, Raufi AG, Safyan RA, Bates SE, Manji GA (2020) BRCA mutations in pancreas Cancer: spectrum, current management, challenges and future prospects. Cancer Manag Res 12:2731–2742 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Younes N, Zayed H (2019) Genetic epidemiology of ovarian cancer in the 22 Arab countries: a systematic review. Gene 684:154–164 [DOI] [PubMed] [Google Scholar]
- Z S. Prevalance of BRCA1 and BRCA2 mutations among breast and ovarian cancer patients in Northern Emirates UAE University; 2020.
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Raw sequencing data for this study were generated at the Dubai Genetics Center, Dubai Health, and are not publicly accessible to ensure compliance with patient confidentiality and privacy regulations. However, the derived data used to support the findings of this study have been included as Supplementary Data file. Data is provided within the manuscript or supplementary information files.