Article Text
PDF
Abstract
Background Pathogenic mutation in BRCA genes causes high cancer risk. Identifying the mutation carriers plays key roles in preventing BRCA mutation-related cancer. Population screening has demonstrated its power for comprehensive identification of the mutation carriers. However, it is only recommended for the Ashkenazi Jewish population with high prevalence of three founder mutations, but not for non-Ashkenazi Jewish populations as the cost-effectiveness could be too low due to their lower mutation prevalence and lack of founder mutation. Population screening would not benefit the majority of the human population for BRCA mutation-related cancer prevention.
Methods We used population BRCA screening in 6000 residents, 1% of the Macau population, an ethnic Chinese population with unique genetic, linguistic and cultural features, and its BRCA mutation has not been analysed before.
Results We called BRCA variants, identified 18 carriers with 14 pathogenic mutations and determined the prevalence of 0.29% in the population (95% CI 0.15% to 0.42%). We compared the testing cost between the Ashkenazi Jewish population, the Sephardi Jewish population and the Macau population, and observed only a few fold differences.
Conclusion Our study shows that testing cost is not the most important factor in considering population BRCA screening, at least for the populations in the developed countries/regions, regardless of the status of mutation prevalence and founder mutation.
- DNA damage
- DNA repair
- early diagnosis
- genetic carrier screening
- genetics
Statistics from Altmetric.com
Introduction
BRCA1 and BRCA2 (BRCA) are essential for maintaining genome stability by repairing double-strand DNA breaks through homologous recombination and non-homologous end joining pathways.1 Pathogenic mutation in BRCA damages the function, leading to genome instability, and increases cancer risk.2 Identification of the mutation carriers plays important roles in preventing BRCA mutation-caused cancers.3 4 Traditionally, this is achieved by referring to family cancer history. However, decades’ practice demonstrates that this approach can only identify no more than half of the carriers, leaving the unidentified carriers missed prevention opportunity until cancer occurs.5 To overcome this obstacle, population screening has been proposed to reach a comprehensive identification of the mutation carriers regardless of familial cancer history.6 The Ashkenazi Jewish population has high prevalence (over 2%) of the three founder mutations.7 Studies in the Ashkenazi Jewish population in Canada, Israeli and London demonstrated that population screening can indeed reach comprehensive detection of mutation carriers in a highly cost-effective manner in the Ashkenazi Jewish population.8–10 As a very promising tool to reach comprehensive identification of BRCA mutation carriers, however, population screening is not recommended for non-Ashkenazi Jewish populations due to the lack of founder mutations, lower mutation prevalence and the presence of many variants of unknown significance (VUSs).10 This would imply that except the Ashkenazi Jewish population of about 10 million, human population would not benefit from population screening and the majority of BRCA mutation carriers would not be identified for prevention of BRCA mutation-related cancer until cancer occurs.6
Of the three factors against the use of population screening in non-Ashkenazi Jewish populations, VUSs are common for all populations, and policy has been well set that VUSs should not be reported.10 No founder mutations and lower mutation prevalence are the major concerns. In contrast to the situation in the Ashkenazi Jewish population, lack of high prevalence founder mutations and lower prevalence in non-Ashkenazi Jewish populations implies that more individuals need to be screened to identify fewer mutation carriers, and the entire BRCA coding region needs to be sequenced. Both factors demand higher screening cost than that in the Ashkenazi Jewish population.9 Population screening certainly requires substantial resources. The question is if the preventive cost in populations with lower mutation prevalence remains significantly lower than the treatment cost. The answer to this question requires detailed information on BRCA variation in a targeted population.
Macau is a specific administrative region in China. As a former Portugal colony, Macau has a population of about 650 000 residents in its land of 30 km2 (figure 1A). About 90% of Macau residents are ethnic Chinese, dominated by Cantonese with distinct genetic, linguistic and cultural characters differing from other ethnic Chinese populations (https://www.dsec.gov.mo/getAttachment/f0477f64-33b6-473c-b4c0-8f9012b41431/C_DEM_FR_2019_Q4.aspx). Macau has an active economy with one of the highest sGross Domestic Product (GDP) per capita in the world. The Macau population is ageing; cancer is the top cause of death and breast cancer is the number one cancer type in Macanese women. BRCA mutation in the Macau population has never been systematically tested. As no BRCA testing system is available in Macau, clinical cases are often referrals to be tested in commercial testing laboratories in nearby Hong Kong. With its moderate population size, ethnicity and wealthy economy, Macau has the potential to eradicate BRCA mutation-related cancers at the population level. In this study, we used the Macau population as a model to experience the use of population BRCA screening in an actual non-Ashkenazi Jewish population. We tested over 6000 Macau individuals, collected BRCA variants, identified pathogenic mutations and determined the prevalence of BRCA pathogenic mutation in the Macau population and discussed the issue of cost-effectiveness.
Geographical location of Macau and BRCA variants between Macau and non-Macau populations. (A) Macau map. (B) BRCA variants between Macau and non-Macau populations represented by BRCA variants in BED, BIC, CIMBA, ClinVar, ENIGMA, LOVE, and UMD databases. BED, BRCA Exchange Database; BIC, Breast Cancer Information Core; CIMBA, Consortium of Investigators of Modifiers of BRCA1/2; ENIGMA, Evidence-based Network for the Interpretation of Germline Mutant Alleles; LOVD, Leiden Open Variation Database; UMD: Universal Mutation Database.
Methods
Sample source and BRCA-targeted sequencing
Leftover blood samples after clinical tests from both men and women over 18 years old were collected from a Macau clinics laboratory. Each donor provided written consent to participate in medical study. All samples were kept as anonymous except gender and age information. The study was approved by University of Macau Institutional Review Board (BESRE17‐APP014‐FHS). The following factors helped to ensure the individuals tested were highly representative of the Macau population and avoided a likelihood of enrichment for genetic cancer susceptibility factors: (1) the clinics laboratory is one of the largest in Macau and located in Macau peninsula, the major resident region in Macau; (2) the sampling process lasted for over 18 months; (3) redundant samples from the same donors at different collecting times were excluded from the study; and (4) patients with cancer were excluded from the study.
Genomic DNA was extracted from each sample using Qiagen Symphony Automated Nucleic Acid Preparation System (Qiagen). BRCA-targeted sequencing was performed for each sample. In the process, BRCA coding region and intron/exon boundaries were amplified by using PCR primers with 349 amplicons (Fluidigm D3 design program, Fluidigm, San Francisco, California, USA); next-generation sequencing (NGS) libraries were constructed by using the targeted DNA Seq Library Preparation kit (Fluidigm LP 192.24 IFC, Fluidigm) and sequenced on HiSeq X Ten instrument with paired end (2×150 bp) with a mean sequencing depth over 4000× (Illumina, San Diego, California, USA).
Data analysis
Sequences were mapped to the human reference genome (hg19) using Burrows-Wheeler Aligner (BWA). SNVs and InDels were called using GATK3.8 by following best practices protocol. Population frequency was determined by referring to Asian population from 1000 Genome Project
Eastern Asian variants, genomAD Eastern Asian variants and ExAC Eastern Asian variants. The variants were further classified into non-synonymous or synonymous, affecting splice junction, and affected stop codon using the Ensembl Variants Effect Predictor program (http://asia.ensembl.org/Tools/VEP). These reference sequences used for variant location were BRCA1: cDNA NM_007294.3, protein NP_009225.1; BRCA2: cDNA NM_000059.3, protein NP_000050.2. The variants were annotated with dbSNP150 through the ANNOVAR program. The variants were further classified into pathogenic, likely pathogenic, Variants of uncertain significance (VUS), likely benign, benign, conflicting interpretation by mapping to ClinVar database. Variants were searched in the following BRCA variation databases to determine their novelty: BIC (https://research.nhgri.nih.gov/bic/), ClinVar (http://www.ncbi.nlm.nih.gov/clinvar/), BRCA Exchange Database (http://brcaexchange.org), BRCA1 and BRCA2 Mutation Database (https://www.aruplab.com/topics/breast-cancer/brcadatabase), Leiden Open Variation Database (LOVD, http://www.lovd.nl/3.0/home) and Evidence-based Network for the Interpretation of Germline Mutant Alleles database database (ENIGMA), Consortium of Investigators of Modifiers of BRCA1/2 (CIMBA, http://cimba.ccge.medschl.cam.ac.uk), and dbBRCA-Chinese (https://dbbrca-chinese.fhs.umac.mo/). The variants classified as pathogenic and likely pathogenic were validated by Sanger sequencing. Classification for those not present in existing BRCA databases was predicted by InterVar.
Results
Collection of BRCA variants
We tested a total of 6314 Macau individuals accounting for about 1% of the Macau population, including 2888 (45.7%) men with an average age of 41 and 3426 (54.3%) females with an average age of 42. We extracted DNA from each individual, performed targeted BRCA sequencing, and identified 659 distinct BRCA variants (table 1 and online supplementary tables 1 and 2).
Supplemental material
Supplemental material
Pathogenic variants identified in Macau population
Features of BRCA variation
On average, each individual carried 12.2 BRCA variants, 6 in BRCA1 and 6.2 in BRCA2. Of the 12 types of variation identified, missense variant had the highest rate of 41.6% (table 1A). Recurrent variants (≥2 carriers) accounted for 50% (53% in BRCA1 and 49% in BRCA2), and the rest ones were singleton present in one individual only. Variants in BRCA2 were 1.5 times more than in BRCA1 (table 1B). The number of BRCA1 variation was 264, 1.5 times lower than the 395 in BRCA2. Thirty-five per cent of Macau BRCA variants were absent in the 22 098 BRCA variants combined from multiple BRCA variation databases, indicating that BRCA variation in the Macau population was highly Macau-specific (figure 1B).
Prevalence of pathogenic variants in Macau population
We classified the variants into pathogenic, likely pathogenic, VUS, likely benign and benign, in which VUS accounted for 22.5% of the total variants. We identified 14 pathogenic variants accounting for 2.1% of total variants (table 1B,C and online supplementary table 3). The number of pathogenic variants in BRCA2 was far more than those in BRCA1 that there were 13 BRCA2 pathogenic variants in 17 individuals but only 1 BRCA1 pathogenic variants in a single individual. All pathogenic variant carriers were heterozygotes. Of the 14 pathogenic variants, 4 were stop–gain; 6 were frameshift indels (4 of frameshift deletion, 2 of frameshift insertion); 1 was splice donor; and 3 were missense mutations. Two BRCA2 variants were located in exon 11 coding for BRC1 (c.3109C>T) and BRC6 (c.5164_5165delAG). c.3109C>T is a founder mutation in Southern Chinese.11 With five carriers of one man and four women and genetic ties of Macau population with Southern Chinese, this mutation is very likely a founder mutation in Macau population. Six of the 13 pathogenic variants were located in exon 11 of BRCA1 and BRCA2, which is the ‘coldspot’ for BRCA pathogenic variants.12 Only one of the 14 Macau BRCA2 pathogenic variants (c.5164_5165delAG) was shared with these in mainland Chinese.13 The presence of 18 pathogenic carriers in 6314 individuals results in the prevalence of 0.29% (95% CI 0.15% to 0.42%), or one carrier in every 345 Macau individuals. Power calculation showed that screening a population size of 6314 individuals at a prevalence of 0.29% provides 98.77% of probability to detect all pathogenic variants in this population. Based on this prevalence, we estimate the existence of 1853 BRCA pathogenic mutation carriers, of which 132 are BRCA1 and 1720 are BRCA2 mutation carriers, in a Macau population of 650 000 (95% CI 0.27% to 0.30%). It is necessary to indicate that NGS was used in our study, which detects mainly the single-base changes and small indels. It is well-known that large structural changes and copy number variations often account for 5%–15% of the spectrum of BRCA variation,14 but these variations could not be effectively detected by NGS. As such, the actual prevalence of BRCA variation would be higher than the observed 0.29% in the Macau population.
Supplemental material
Cost-effective calculation
The prevalence of 0.29% in Macau population is seven times lower than the 2% in Ashkenazi Jewish population. The cost of testing varies in different places. It was about 200 Macanese pataca (MOP (10£, 10 MOP=1£) per case in our study, and the total cost for the 6314 cases was about 1 260 000 MOP (£126 000). With the identified 18 pathogenic mutation carriers in this population, the cost of identifying the carriers was 70 000 MOP (£7000) per mutation carrier. Testing at the same scale in Ashkenazi Jewish population with a prevalence of 2% would identify 126 pathogenic mutation carriers at the cost of 10 000 MOP (£1000) per carrier. The prevalence is 0.7% in the Sephardi Jewish population, carrying only one of the three Jewish founder mutations (BRCA1 185delAG/ c.68_69delAG).15 Testing Sephardi Jewish population would identify 44 pathogenic mutation carriers at the cost 28 636 MOP (£2864) per carrier. Therefore, the cost of testing in the Macau population would be sevenfold higher than in the Ashkenazi Jewish population and 2.4-fold higher than in the Sephardi Jewish population.
Discussion
Genetic testing is increasingly applied in medical practice as best exemplified in prevention and treatment of BRCA mutation-related cancer, and population screening has been shown as a powerful tool for prevention of genetic diseases at the population level. For example, Tay-Sachs disease in Israeli Jewish population has been successfully eradicated through newborn screening.16 Population BRCA screening has demonstrated its highly cost-effective in Ashkenazi Jewish population with higher prevalence of the three founder mutations.8–10 Population screening still maintains highly cost-effective in Sephardi Jewish at 0.7% prevalence.14 Therefore, population screening is recommended for all the Jewish populations.14 Our study shows the cost of testing in Macau population was sevenfold higher than that in the Ashkenazi Jewish population and over 2.4-fold higher than that in the Sephardi Jewish population. Within a range of a few fold differences of screening cost, it is difficult to set an artificial cut-off to determine if population screening is or is not cost-effective. Another important factor to be considered is that the costly Sanger sequencing was used in detecting the founder mutations in Jewish population,9 whereas the cheaper NGS is used in testing the entire BRCA coding region in non-Jewish population.13 From the detailed analyses in Ashkenazi Jewish and Sephardi Jewish populations and our current study in the Macau population, it is likely that population BRCA screening would always be cost-effective over family cancer history-based approach regardless of the status of prevalence and founder BRCA pathogenic mutations in different human ethnic populations. It is necessary to indicate that we only used sequencing cost as the reference for the comparison. However, this is simplistic as the actual cost can be much higher than sequencing cost only, such as infrastructure, testing uptake rate, accessibility to effective screening and genetic counselling for preventive strategies.
The estimated prevalence of BRCA pathogenic mutations varies in different populations, such as 0.18% in Malaysian,17 0.26% in Japanese,18 0.38% in Han Chinese,13 0.38% in Mexicans19 and 0.53% in American.20 The decision for applying population BRCA screening would likely be made more by non-cost factors of social, political, public interests and patient benefits. This is best represented for the situation in Macau that with its wealthy economy, the cost would not be an important issue in planning population BRCA screening in the Macau population. Nevertheless, population BRCA screening, if widely applied in non-Ashkenazi ethnic populations, should greatly promote the prevention of BRCA mutation-related cancer, the type of cancer mostly preventable among all cancers.
Ethics statements
References
Footnotes
Contributors ZQ: sample collection, data collection, annotation and analysis; CNK, LJ, ZL, HKL, LW and GM: sample collection; HD: data collection and analysis; MQ: statistics data analysis; SMW: funding, experimental design, data analysis and interpretation, and manuscript writing.
Funding This work was funded by a Macau Science and Technology Development - Ministry of Science and Technology of People's Republic of China fund (0077/2019/AMJ), grants from the University of Macau (SRG2017-00097-FHS, MYRG2019-00018-FHS), a grant (FHSIG/SW/0007/2020P) and a startup fund from the Faculty of Health Sciences, University of Macau (SMW). We are thankful to the Information and Communication Technology Office, University of Macau, for providing the high-performance computing cluster resource and facilities for the study.
Map disclaimer The depiction of boundaries on this map does not imply the expression of any opinion whatsoever on the part of BMJ (or any member of its group) concerning the legal status of any country, territory, jurisdiction or area or of its authorities. This map is provided without any warranty of any kind, either express or implied.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.