To the editor

Breast cancer incidence in Asia has doubled over the past 10 years and is one of the highest in Hong Kong [1]. Studies in western populations have found differences in breast cancer epidemiology between Asian and Caucasian populations, potentially due to interactions between different lifestyle and genetic characteristics [2–4]. As the genetic predisposition to breast cancer is increasingly understood, it has been suggested that there are differences in different ethnic populations. To date, only a few investigations on the spectrum of BRCA [5, 6] mutations in Chinese populations have been performed, most of which were performed in single institutions or a small number of medical centers [7–9]. Some studied only the BRCA1 gene [10, 11] and most did not perform mutational screening using conventional full gene sequencing [12, 13].

Between March 2007 and February 2008, patients with breast and/or ovarian cancer were recruited from the Hong Kong Hereditary Breast Cancer Family Registry, which was formed in March 2007. In this Registry, all patients were recruited from 11 government-based hospitals, private hospitals, and clinics that cover all three areas of Hong Kong. Sequence analysis of all exons in BRCA1 and BRCA2 of a total of 119 high-risk female probands were conducted. The mean age at diagnosis of breast cancer was 41.6 years (range 23–73) and that of ovarian cancer was 43.8 years (range 23–65). All probands were from Chinese ancestry and 81.5% were from Guangdong province of southern China. Our sequencing results identified 22 (18.5%) deleterious BRCA gene mutations. A total of 8 (8/119, 6.7%) gene mutations were in BRCA1 and 14 (14/119, 11.8%) in BRCA2. The spectra of mutations are illustrated in Table 1. Among those mutations detected, only one BRCA1 (c.4987_5277del) and one BRCA2 (c.8633_8755del) deletion mutation were found by multiplex ligation-dependent probe amplification (MLPA) and confirmed by DNA sequencing.

Table 1 Spectrum of BRCA pathogenic mutations identified
Full size table

Of the 22 deleterious mutations found, seven (31.8%) were new mutations, which have not been published in the Breast Information Core Database of National Institute of Health (BIC) as described in Table 1. All the seven new mutations cause sequence frame-shift, leading to early termination of translation for protein products. Of these new mutations, four were in BRCA2 and three were in BRCA1. Reports have found a high frequency of genetic sequence variants in different ethnic populations [8, 11, 14, 15]. This is likely due to the limited knowledge of the mutation spectrum in different ethnic populations, where BRCA testing has not been widely performed [16]. Since we have identified a number of missense mutations and synonymous polymorphisms of unknown significance, further functional analysis and disease risk association studies are needed to be performed, in the future, to characterize the clinical significance of these mutations.

One recurrent BRCA2 mutation was found in four probands in this cohort and accounted for 28.6% of BRCA2 mutations and 18.2% of all BRCA mutations. The recurrent mutation c.3109C>T; p.Gln1037X in exon 11 of BRCA 2 was found in four unrelated probands in this cohort. We have also seen this recurrent mutation in two other independent Hong Kong Chinese male breast cancer patients from another patient cohort. Accordingly, this recurrent mutation was identified in a total of six independent individuals (proband# 45, 57, 91, 118, and male patient# M1, M2). This mutation resulted in the creation of an early termination codon and a truncated BRCA2 protein product. In order to confirm the founder effect, haplotype analysis using eight microsatellite polymorphic markers (D13S1304, D13S217, D13S289, D13S1699, D13S1698, D13S171, D13S1695, and D13S267) that are in close proximity to BRCA2 was performed in the 6 index cases and 8 family members from 3 index cases and 50 unrelated Chinese individuals. A common haplotype was found to co-segregate together with the mutation, but not found in the mutation negative family members of the probands. Based on the haplotype analysis of our control cohort of 50 unrelated Chinese individuals, we estimated the random population frequency of the shared haplotype is ~1.13 in 10,000. Recombination event is observed in two of the six individuals for markers D13S217 and D13S267 which are susceptible for recombination events, as their recombination frequency are ~3.53 and 1.36% respectively. The haplotype analysis confirmed the mutation c.3109C>T, a founder mutation of southern Chinese.

Variations in techniques of mutation detection can result in significant inconsistencies between studies [3, 17, 18]. In addition, large-scale rearrangements in BRCA1 have been reported to be as frequent as 2–12% in high risk families [19, 20]. Among all BRCA1 mutations, 7–40% may be due to large rearrangements [21, 22] and similarly in 2–8% of BRCA2 families [23–25]. Recent studies have found that without analysis of genomic rearrangements some mutations may be missed [20, 26]. There is increasing data suggesting that large genomic alterations may be more common in specific ethnic populations [21, 27]. We have chosen to perform MLPA screening in all our probands to improve the under reporting of BRCA1/2 mutations, where knowledge of Chinese race is already limited. Indeed one (4.8%) BRCA1 and 1 (4.8%) BRCA2 large deletion were found only by MLPA, and overall identification rate of 9.5% of all the BRCA mutations was found in the probands of this cohort. Hence, although conventional DNA sequencing is laborious and costly for such large size of BRCA1 and BRCA2, we believe that the combination of conventional sequencing with MLPA is currently one of the best choices for evaluating a wide range of genomic alterations in BRCA1 and BRCA2 genes.

In conclusion, 18.5% of our high risk Chinese cohort carries the BRCA mutation. We identified seven (7/22, 31.8%) new deleterious mutations and also confirmed and reported a BRCA2 founder mutation in our Hong Kong Chinese cohort, which mainly consist of southern Chinese. Knowing the frequency and range of mutations in a specific ethnic population provides important factor in genetic counseling and cancer risk assessment. The discovery of this founder mutation may provide a cost-effective option to screen southern Chinese population.