Prevalence and spectrum of germline BRCA1 and BRCA2 mutations in multiethnic cohort of breast cancer patients in Brunei Darussalam

Introduction

Breast cancer is the most common cancer affecting women worldwide with varying incidence rates across the world [1]. Among the breast cancer risk factors, genetic predisposition confers the highest risk in breast cancer progression [2]. About 5-10% of breast cancers are inherited with approximately 30% of the inherited breast cancers were attributed to germline mutations in high penetrance breast cancer susceptibility genes, Breast Cancer susceptibility genes type 1 (BRCA1) [3]and Breast Cancer susceptibility genes type 2 (BRCA2) [4]. BRCA1 and BRCA2 are tumour suppressor genes (TSGs) that function by suppressing the growth of tumour cells via multiple DNA damage and repair pathways in the cells [5–10]. Notably, the BRCA1 protein is multi-functional as it also regulates cell cycle through a number of mechanisms [5,6,8,10–15].

BRCA1 and BRCA2 germline mutation carriers had an increased cumulative risk of 72% and 69%, respectively to develop breast cancer by 80 years old [16]. BRCA1 and BRCA2 carriers would also have an increased risk of developing ovarian, prostate, and pancreatic cancers [16]. Breast cancer patients who inherit germline mutations in these genes, in particular BRCA1 carriers, are commonly associated with triple negative breast cancer (TNBC), diagnosed at an early age of onset (≤40 years), diagnosed with bilateral breast cancer, having family history of breast and/or ovarian cancers in the first- and second-degree relatives, and having an ovarian cancer [17–21]. The discovery of these two genes being linked to breast and ovarian cancers has led to the increased importance of genetic testing where continuous research and development had ultimately resulted in BRCA1 and BRCA2 genetic tests available at a more affordable cost and with a higher sensitivity and specificity [22]. Moreover, increased awareness on available personalised treatment for affected patients such as olaparib, an oral poly ADP ribose polymerase (PARP) inhibitor, which has been shown to provide a significant benefit over standard therapy among the BRCA mutation-carrier patients [23–24], has led to the increasing demand for rapid BRCA testing preferably at the first diagnosis [22]. This rapid increase caused pressure for diagnostic laboratories to provide a genetic test with a shorter turnaround time [22]. The two most commonly used platform for DNA sequencing are Sanger sequencing and Next-generation sequencing (NGS). Although NGS has largely overtaken Sanger sequencing due to its cost-effective ability in screening a larger set of samples in parallel and simultaneous screening of multiple cancer susceptibility genes in one sample, Sanger sequencing is still used in laboratories today when the main objective is to screen a single gene only.

Brunei Darussalam is a small country (5,765 sq km) on the Borneo Island, located in the South-east Asia (SEA) region, bordering the South China Sea and East Malaysia [25]. The population is estimated over 400,000, comprises 66% Malay, 10% Chinese, 3% other indigenous and 21% other ethnic groups [25]. Breast cancer was one of the top three leading causes of cancer mortality among women in Brunei Darussalam with an incidence rate (age-standardized rate [ASR]) of 55.9 per 100,000 women which is the second highest among countries in the SEA region, but considerably lower compared to the Western Europe (ASR, 90.7) [1,26]. Notably, the incidence rate of breast cancer in Brunei varies by ethnicity, with the highest rate observed in Chinese (ASR, 60.4) [27], followed by Malays (ASR, 48.1) [27], and Others (ASR, 12.3) [28]. There has not been any formal study conducted on finding the contribution of genetic and non-genetic factors in the rising incidence of breast cancer in Brunei. Therefore, studying various aspects of breast cancer in the Brunei patients could help to understand and determine the probable cause that leads to the rise in the incidence of the disease in the population, and in planning a better health care for the Brunei population. The contribution of genetics in breast cancer specifically in the involvement of susceptibility genes has been continuously researched on in the Western and developed Asian countries resulting in many reports on the spectrum of variants within the susceptibility genes from all over the world. In the SEA region, the mutation spectrum of BRCA1 and BRCA2 genes in breast cancer has been studied in Singaporean [18,19,29–31], Malaysian [21,32–37], Filipino [38,39], Vietnamese [40,41], Thai [42,43], and Indonesian [44,45] populations leading to the discovery of novel pathogenic variants from different ethnicities, suggesting the mutation spectrum of the BRCA1 and BRCA2 genes are still understudied. Furthermore, a number of the variants identified in the SEA breast cancer populations has also been found in other populations such as the African, European and American suggesting that they are not specific to the reported population, implying the genetic heterogeneity and randomness of the mutations. Interestingly, most of the variants identified in the breast cancer population in East Malaysia which is also located on the Borneo Island [33] were not identified in the West Malaysia population suggesting a difference in population genetics between East and West Malaysia.

Some of the variants identified in the Filipino breast cancer population who mostly were a combination of Malay and other population ancestry have also been proposed to be founder mutations due to the shared haplotype markers [39]. Currently, there are no genetic mutation data that report on the frequency of any genes related to breast cancer in Bruneian patients with or without family history. Thus, genetic contribution in breast cancer and other related cancers in Bruneian population remains uninvestigated. In this study, we investigated a population-based unselected series of Brunei breast cancer patients to determine the prevalence and spectrum of BRCA1 and BRCA2 mutations using Sanger and next-generation sequencing methods, and assess their association with sociodemographic, clinicopathological, and family history characteristics of the population.

Materials and methods

Results

This was the first study of its kind in the Brunei breast cancer patients. Of the 120 recruited breast cancer patients, 94 (78.3%) were Malays, 19 (15.8%) were Chinese, and 7 (5.8%) were others. The mean age at diagnosis was 49.1±10.51 years, with peak of breast cancer incidence occurred in the 51-60 years old age group, and one (0.8%) was a male breast cancer (Table 1).

Seven (5.8%), two (1.7%), one (0.8%), and one (0.8%) patients had personal history of bilateral breast cancers, ovarian cancer, endometrium cancer, and liver cancer, respectively. Assessment of family history of breast and/or ovarian cancers was made based on whether the affected family members of the patients were their first and/or second-degree relatives. Of the 120 cases, 36 (30.0%) had family history of breast and/or ovarian cancers in their first and/or second-degree relatives while 47 (39.2%) had no family history of cancers at all. Of patients with family history of breast cancer, 17 (14.2%) had one first-degree family member affected with breast cancer, and 3 (2.5%) had two first-degree relatives affected with breast cancer. For patients with family history of ovarian cancer, 5 (4.2%) and 2 (1.7%) had only one affected member in the first- and second-degree relatives respectively (Table 1).

BRCA1 and BRCA2 mutations

Of the 95 variants identified in the study population, 29 (30.5%) and 66 (69.5%) were BRCA1 and BRCA2 variants respectively (S1 – S4 Tables). Of the 29BRCA1 and 66BRCA2 variants, 11 (37.9%) and 22 (33.3%) were novel respectively (S2– S4 Tables). No large rearrangements were detected in this study.

Prevalence of pathogenic mutations and VUS

Our study identified three frameshift BRCA2 deleterious mutations in 5 patients which led to the introduction of premature stop codon in the resulting BRCA2 protein (S1 Table, Fig 1 and S2 Fig), and one likely pathogenic BRCA2 missense mutation in one patient (S2 Table and Fig 1). Therefore, the overall prevalence of germline BRCA1 and BRCA2 carriers among Brunei breast cancer patients were 0% and 5%, respectively. All of the identified pathogenic and likely pathogenic mutations have been reported, and were rare in the general population (<0.05% in gnomAD and 1000 Genome). One of the BRCA2 deleterious mutations (c.5164_5165delAG) appeared to be a recurrent mutation.

• Download figure
• Open in new tab

Fig 1. Schematic presentation of identified BRCA2 pathogenic mutations among Brunei breast cancer cases.

Excluding one patient with simultaneous germline pathogenic mutations, our study also identified 9BRCA1 and 32 BRCA2 VUS in 10and 31 patients, respectively (S3 Table). Thus, the prevalence of BRCA1 and BRCA2 VUS carriers among Brunei breast cancer patients were 8.3% and 25.8%, respectively. Four of the BRCA1 VUS and 14 of the BRCA2 VUS were novel while the others have been reported. Almost all of the reported VUS have no recorded data in the general population (S3 Table) while those that with data were found to be rare in the general population (<0.05% in gnomAD and 1000 Genome).

Characteristics of Brunei’s germline pathogenic BRCA carriers

S5 Table shows the characteristics of the Brunei breast cancer patients with deleterious and damaging missense mutations. Collectively, one (16.7%) of the carriers was male while the rest were female. Two (33.3%) of the affected were diagnosed with bilateral breast cancer. The molecular subtype of the first breast cancer diagnosis showed that one (16.7%) patient was TNBC, one (16.7%) was Luminal A, while the rest were Luminal B. Two (33.3%) of the carriers had an age at onset of ≤40 years while the others were ≥40 years, and one (16.7%) of the carriers had no family history of cancers at all. The proportion of our BRCA1/2 carriers by ethnicity was 3/19 (15.8%) and 3/94 (3.2%) in Chinese and Malay, respectively.

There is no significant difference observed between the overall median age of diagnosis among carriers (48 years) and non-carriers (49 years) (p=0.609, Table 2). Brunei BRCA2 carriers were found to be more likely to have family history of breast and/or ovarian cancer, and have≥1 affected family members in the first-degree with breast cancer (p = 0.027 and p = 0.001 respectively, Table 2). There are no significant association between Brunei BRCA2 carriers and other selected clinical characteristics in Table 2.

View this table:

• View inline
• View popup

Table 2. Association analysis of Brunei breast cancer patients with pathogenic mutation with selected clinical characteristics.

Discussion

Prior to introducing cancer genetic test in a community, genetic variation profile of the said community must first be known to have an overview of the contribution of genetics in the population. Although cancer incidence and mortality in Brunei Darussalam have been increasing for the past 10 years, there are still no data available on the proportion of Brunei cancer patients carrying germline or somatic mutations in their cancer predisposing gene. BRCA1 and BRCA2 gene frequencies in breast and ovarian cancer patients in other East-Asian populations have been continuously researched on which had led to the discovery of potential founder mutations [29,39,55,65]. Notably, identifying who should be offered cancer genetic test in the Asian community remains challenging as the currently available guidelines for testing have been shown to omit at least 20% of patients carrying BRCA mutations from being tested [21], suggesting the testing criteria on BRCA1 and BRCA2 germline mutations which have been well established in the Caucasian population were not readily applicable to the Asian population.

Our study, which is the first genetic study of its kind in Brunei, showed that the overall combined frequency of germline pathogenic BRCA1 and BRCA2 mutation carriers among 120 recruited Brunei’s breast cancer patients was 5% (n = 6), which is lower than those reported at 12.3% in women of European descendent [66]. Our data are relatively similar to previously reported prevalence from other Asian countries in particular from the South-East Asian region which is 4.7% in Malaysia [21] and 5.1% in the Philippines [39] except for Singapore [19] and Vietnam [40] which were reported at 17.4% and 0.8%, respectively. Interestingly, other Asian countries outside of the South-East Asian region had reported a relatively higher prevalence of BRCA1 and BRCA2 mutations which was almost similar (or higher for some) to the findings reported in European women with14.0% in Korea [67] and 24.7% in Pakistan [68]. Japan [69], Hong Kong [54,55] and China [20] however, reported a prevalence of 4.2%, 7.9% to 8.8%, and 8.3%, respectively. Comparatively, all these countries had reported prevalence of germline mutations in both BRCA1 and BRCA2 genes while all germline BRCA mutations identified in our study were attributed to BRCA2 gene only.Our findings were consistent with those from all the aforementioned Asian countries where most of the germline pathogenic BRCA mutations in the Brunei breast cancer patients were identified in the BRCA2 gene compared to BRCA1 which was reported to be more affected in the European women [66].

Our study showed that one recurrent deleterious frameshift mutation, BRCA2c.5164_5165delAG (S1722Yfs*1725), was identified in 3 (2.5%) patients of Chinese ethnicity. In particular two of the affected patients, 4G-004 and 4G-003, were related to each other (mother and daughter, respectively). This mutation was suggested to have a founder effect in the Southern China province population. While it was unknown whether our Chinese index cases ancestry were originally from the Southern China, there was still a probability that the occurrence of this mutation might be linked to a common ancestral origin. Moreover, this mutation has also been reported in the Chinese ethnicity in Malaysia [34] and Singapore [18] which further support the probability of the founder mutation theory. However, it is acknowledged that the probability could be slightly lowered due to the fact that two out of three of the carriers of this mutation were relatives where the variant could have been inherited from their maternal side. Although the other two deleterious frameshiftBRCA2 mutations identified in this study only occur in one (0.8%) of the study population, the mutations c.3170_3174delAGAAA (K1057Tfs*1064) and c.1763_1766delATAAA (N588Sfs*612) had each been reported as a founder mutation in the French-Canadian [52]and Colombian[51] populations respectively. The probability of these two mutations being linked to the two populations was highly unlikely as the carriers were both of Malay ethnicity, and the family history from both carriers did not support a common ancestral origin with the aforementioned populations. Interestingly, in the South-East Asia (SEA) population, the BRCA2 mutation c.1763_1766delATAA (N588Sfs*612) was only ever reported in a breast cancer patient in Sarawak, Malaysia who was diagnosed <40 years old with triple negative breast cancer and family history of breast cancer [33].This observation suggests that all populations have different population genetic predisposition towards breast cancer. Pathogenic mutations determined as a founder effect in one population may only be detected in a small frequency in other populations of different ancestral origin. Moreover, for the two patients identified with these mutations, there is a 50% chance that their family members carried the same BRCA2 mutation that could result in the development of breast cancer in these family members.

The three aforementioned frameshift mutations introduced premature stop codon which was known to be associated with an increased risk of breast cancer. The effect of frameshift mutations resulting in the stop codon has been shown to destabilise BRCA mRNA leading to a haplo-insufficiency resulting in allelic imbalance [70]. This in turn caused a significant reduction in the expression ratio between the mutant and the wild-type alleles [70]. The allelic imbalance resulted in a decreased level of both BRCA transcripts and proteins contributing to the increased risk of breast cancer [70].The resulting BRCA2 truncated protein from the three identified pathogenic variants in our study showed a loss of important domains, including: (1) the BRC repeats domain (all for c.1763_1766delATAAA, BRC-2 to BRC-8 for c.3170_3174delAGAAA, BRC-6 to BRC-8 for c.5164_5165delAG); (2) the DNA binding domain (DBD); and (3) the two nuclear localisation signals (NLS) in the C-terminus region (S2 Fig).The highly conserved eight BRC repeat domains and the DNA binding domains in the BRCA2 protein mediate the interaction of ssDNA and/or dsDNA with RAD51 molecules in the homologous recombination (HR) regulated DNA repair pathway. In HR regulated pathway, BRCA2 is localised onto sites of dsDNA break by forming complex with other proteins including BRCA1 and PALB2, before it mediates the loading of RAD51 onto the ssDNA and initiate the repair. Loss of these RAD51-interacting domains in the BRCA2 protein prevent DNA repair in error-free HR regulated pathway, allowing the error-prone non-homologous end joining (NHEJ) pathway to take over. Moreover, it has been demonstrated that the deletions of all RAD51-interacting domains cause embryonic lethality in mice, but deletion of several BRC repeats motif showed a less severe phenotype although it was observed that the mice died very young from cancer development [71]. The DBD in BRCA2 was also involved in transcriptional activity where it bound ssDNA before drawing histone modifiers which led to the initiation of the transcriptional process [10]. The loss of the two NLS in the C-terminus region due to the truncation was reported to result in the mutant BRCA2 being cytoplasmic since it could not be translocated into the nucleus, hence causing it to become non-functional [72].

One likely pathogenic missense mutation was identified in the only patient who have no family history of cancers at all (S5 Table). BRCA2 mutation c.8524C>T (R2842C) was located in the BRCA2 DNA binding domain suggesting it can affect the protein function. An assessment made on the clinical relevance of VUS identified in the BRCA2 DBD by using a validated functional assay of BRCA2 homologous recombination (HR) DNA-repair activity had showed the mutation R2842C had intermediate or partial effects on BRCA2 function [57]. Another functional assessment of human BRCA2 variants using a mouse embryonic stem cell (mESC)-based assay supported that although the mutation R2842C is able to complement the loss of cell viability, its capacity to perform BRCA2 function i.e., homology directed repair (HDR) showed more than 50% reduction in HDR capacity compared with wild-type BRCA2 expressing cells[58,59].These results suggested that the mutation R2842C is characterized as a hypomorphic variant in which the studies concluded that the mutation is intermediate between pathogenic and likely pathogenic variants [58,59]. In our study, the effect of thehypomorphicBRCA2 germline mutation R2842C was observed in patient 2G-022, in which the patient had an increased susceptibility to familial breast and/or ovarian cancers even though she did not have any family history of cancers at all. Notably, none of the founder mutations found in the SEA population were identified in our study population.

Similar to other genetic variation analysis conducted in other populations, 41 (43.2%) of the identified variants in the study population were classified as VUS (S3 Table). The overall combined frequency of patients with BRCA1 and BRCA2 VUS in the study was 35 (29.2%), in which 6 patients have both BRCA1 and BRCA2 VUS. Our findings are higher than those reported in other populations, such as 5.7% in European [66], and 1.6% in Malaysian [21] women, respectively. It has been shown that the prevalence of VUS in a study population was inversely correlated with the total number of individuals tested [66]. Hence, if our total study cohort population was higher than 120, our prevalence could be relatively similar with other countries’ findings as demonstrated in three other Asian countries’ studies conducted comprising over 2500 breast cancer patients; namely China [20], Malaysia [21], and Japan [69].

Collectively, Brunei BRCA2 carriers exhibited almost the same phenotypic association pattern as other BRCA2 carriers in China [20] and Malaysia [21], where Brunei BRCA2 carriers were found to be more likely to have family history of breast and/or ovarian cancers, and having more first-degree relatives affected with breast cancer. Consistent with other studies, our findings showed that TNBC is not significantly associated with BRCA2 carriers[17,19,21].

It is acknowledged that this study is limited by the small size of the study population even though it was the first and largest cohort study for genetic testing in Brunei. Future study on a larger cohort of study population could confirm our findings on the prevalence of germline mutations in the Brunei breast cancer population. Moreover, it could also provide a more comprehensive data for phenotypic characteristics of predisposing carriers in our breast cancer population attributed to each gene. Our current study showed that none of our study population tested for BRCA1 and BRCA2 large rearrangements showed positive for the test. Therefore, there are no data on the contribution of large rearrangement of BRCA1and BRCA2 in the Brunei breast cancer population presented in this thesis. Given the constraint in getting resources for performing large rearrangements analysis in some of the study population (Batch 2), there is an approximately 10% probability that some carriers may have been missed in this study [19].

Conclusions

The findings from this study have highlighted the contribution of genetics, specifically BRCA1 and BRCA2 genes in 120 unselected series of Brunei breast cancer population. The prevalence of germline BRCA1 and BRCA2 mutation carriers among Brunei breast cancer patients (5%) is similar to that of other Asian populations confirming the difference in contribution of BRCA1 and BRCA2 in both the Caucasian and Asian populations. One BRCA2 mutation observed only in Brunei breast cancer population of Chinese ethnicity contributes to the probability of the mutation being a founder effect in the Southern Chinese population. Future studies should investigate the contribution of germline mutations of BRCA1 and BRCA2 genes in a larger case-control cohort study of Brunei breast cancer population to confirm our findings.

Supporting information

S1 Fig. The study strategy to detect BRCA1 and BRCA2 mutations.

S2 Fig. Schematic presentation of truncated BRCA2 protein in the study population. The three identified BRCA2 pathogenic mutations, c.1763_1766delATAAA, c.3170_3174delAGAAA and c.5164_5165delAG led to BRCA2 protein truncation at amino acids positions 612, 1064, and 1725 respectively. The truncation caused the loss of important domains affecting the function of BRCA2 predominantly in the HR-regulated DNA repair pathway.

S1 Table. Deleterious mutations identified in the Brunei breast cancer patients.*, Termination; AA, amino acid; AF, Allele Frequency; Freq, Frequency; FS, Frameshift; Nov, Novel; Ref, Reference; Rep, Reported.

S2 Table. Damaging missense variants identified in the Brunei breast cancer patients. AA, amino acid; AF, Allele Frequency; B, Benign; D, Damaging; Freq, Frequency; N, Neutral; NA, Not Available; Nov, Novel; PrD, Probably Damaging; Ref, Reference; Rep, Reported; T, Tolerated.

S3 Table.Missense variants identified in the Brunei breast cancer patients. AA, amino acid; AF, Allele Frequency; B, Benign; D, Damaging; Freq, Frequency; N, Neutral; NA, Not Available; Nov, Novel; PoD, Possibly Damaging; PrD, Probably Damaging; Ref, Reference; Rep, Reported; T, Tolerated; VUS, Variant of uncertain significance.

S4 Table. Synonymous variants identified in the Brunei breast cancer patients. AA, amino acid; AF, Allele Frequency; Freq, Frequency; NA, Not Available; Nov, Novel; Rep, Reported.

S5 Table. Characteristics of the Brunei breast cancer patients with deleterious and damaging missense mutations.+, Positive; -, Negative; ‘XX, year diagnosed or deceased; AA, Amino acid; Aff, Affected; Au, Aunt; Bil., Bilateral; Br, Breast; Bro, Brother; Ca, Cancer; Chi, Chinese; Cou, Cousin; Dau, Daughter; de., deceased; dx., diagnosed; DC, Ductal carcinoma; DCIS, Ductal carcinoma in situ; ER, Oestrogen receptor; Eth, Ethnicity; F, Female; Fa, Father; Gen, Gender; GrMo, Grandmother; Her2, Human Epidermal Growth Receptor-2; IDC, Invasive ductal carcinoma; ILC, Invasive lobular carcinoma; L, Left; Liv, Liver; Lu, Lung; M, Male; Mal, Malay; Mat, Maternal; Mo, Mother; O, Older; Ov, Ovarian; Pat, Paternal; PR, Progesterone receptor; R, Right; Rec, Rectal; Sis, Sister; Sync, Synchronous; Uni, Unilateral; Y, Younger; yr., years.