Abstract
Objective
Breast cancer (BC) is the most common form of cancer among Asian females. Mutations in the BRCA1/BRCA2 genes are often observed in BC cases and largely increase the lifetime risk of having BC. Because of the paucity of high-quality data on the molecular spectrum of BRCA mutations in South Asian populations, we aimed to explore these mutations among South Asian countries.
Methods
A systematic literature search was performed for the BRCA1 and BRCA2 gene mutation spectrum using electronic databases such as PubMed, EMBASE, and Google Scholar. Twenty studies were selected based on specific inclusion and exclusion criteria.
Results
The 185delAG (c.68_69del) mutation in exon 2 of BRCA1 was the most common recurrent mutation and founder mutation found. Various intronic variants, variants of unknown significance, large genomic rearrangements, and polymorphisms were also described in some studies.
Conclusions
The South Asian population has a wide variety of genetic mutations of BRCA1 and BRCA2 that differ according to countries and ethnicities. A stronger knowledge of various population-specific mutations in these cancer susceptibility genes can help provide efficient strategies for genetic testing.
Introduction
BRCA1 and BRCA2 are the two human breast and ovarian cancer susceptibility genes discovered in 1994 and 1995, respectively.1,2 The BRCA1 and BRCA2 genes are located on chromosome 17q and 13q, respectively, and encode factors involved in cell cycle control, gene transcriptional regulation, DNA damage repair, apoptosis, and various other vital cellular processes. 3
Globally, breast cancer (BC) is one of the most common cancers. According to global surveys, the overall rate of BC is higher in American and European women compared with Asians, which may be related to the lifestyle of the Asian population. 4 Though significant differences in incidence and mortality rates between developing and developed countries have been observed, BC comprises a leading cause of death among women worldwide. 5 In addition, BC is the most common cancer among Asian females. 6 BRCA1 and BRCA2 mutations are observed in approximately 30% of all hereditary or genetic BC cases. 7 In contrast to western countries where BC incidence peaks among postmenopausal women in their 60s, peaks of BC incidence in Asian countries are observed among premenopausal women in their 40s. 8
BRCA1 and BRCA2 are clinically significant because deleterious mutations in these genes are correlated with an increased lifetime risk of BC as high as 60% to 85%, often among people with a positive family history.9–12 In addition, women with BRCA1 mutations have an increased risk of developing ovarian cancer, while men are at increased risk of developing prostate cancer. 13 Furthermore, carriers of BRCA2 mutations are at increased risk of gallbladder, bile duct, and stomach cancers, as well as melanoma. 14
The prevalence and distribution of BRCA1/BRCA2 mutations are variable because of population-specific recurrent or founder mutations. Accurate identification of the population-specific mutation spectrum is the foremost and most crucial step for assessing cancer risks, applying the importance of genetic testing into clinical practice, establishing preventive measures, and planning for cancer management strategies. 15
The developing countries of South Asia inhabit approximately 588 million women over 15 years of age who are at risk of ever-increasing incidence of BC.16,17 A lack of awareness and screening protocols, very few diagnostics centers with limited or no access, and lower health care standards have resulted in a higher BC patient mortality rate in developing countries.18,19 Moreover, there is a paucity of high-quality data on the epidemiology, biology, and environmental background of BRCA mutations in South Asian populations, and the central cancer registries with detailed nationwide data on BC mutations are also lacking in these countries. 20 These necessitate a rigorous review of BRCA1/BRCA2 mutations in this region. Thus, this study was carried out to identify the frequency of BRCA1 and BRCA2 gene mutations and explore the spectrum of these mutations among the South Asian population.
Methods
Search strategy
A systematic literature search of the BRCA1 and BRCA2 gene mutation spectrum was performed from 10 September to 20 September 2020 using electronic databases like PubMed, EMBASE, and Google Scholar. The electronic databases were searched for peer-reviewed articles published from 1 January 2000 to 1 September 2020. “Breast Neoplasm”, “Hereditary breast and ovarian cancer syndrome”, “Genes, BRCA1”, “Genes, BRCA2” were the Mesh terms used, and “Breast cancer”,” Hereditary breast and ovarian cancer”, “BRCA1 mutation” and “BRCA2 mutation” were the relevant keywords used, along with the names of all the South Asian countries (Afghanistan, Bangladesh, Bhutan, India, Pakistan, Maldives, Nepal, and Srilanka) connected with Boolean operators “OR” and “AND” wherever appropriate. The search details are further mentioned in Supplementary File 1. Additionally, rigorous manual searches of the reference section of the included articles and the relevant review articles were conducted.
Eligibility criteria
All the peer-reviewed articles published in English, including information on the spectrum of either BRCA1 or BRCA2 gene mutations in South Asian women with any form of BC alone or both breast and ovarian cancers, were considered eligible for inclusion in this study. Studies conducted in the South Asian region were also included.
The following exclusion criteria were followed:
Studies with insufficient information about a mutation in BRCA1 and BRCA2; Review articles and research protocols; Case series/case reports; Symposium/conference proceedings, commentaries, editorials, letters, views, and opinions; Studies with unclear study designs and unavailable data for risk calculation; Full text unavailable; Articles not in the English language.
For two or more studies that included the same set of patients, we included the study with the larger number of patients.
Data extraction and analysis
Two independent authors (SK and SS) rigorously reviewed the selected studies that met our inclusion criteria and extracted the precise information on headings: author(s), year of publication, number of patients, age at diagnosis, country of origin, clinical phenotype, BC type, method of mutation detection, exon/intron, nucleotide change, amino acid change, mutation type, frequency, and founder mutations. This information was recorded in Microsoft Excel 2013 (Microsoft Corporation, Redmond, WA, USA) (Supplementary Tables S1 and S2). Mutations were classified into different groups according to the original classification the authors used during the study period. All mutations collected were reviewed through the Breast Cancer Information Core database (cBIC) (https://research.nhgri.nih.gov/projects/bic/), Leiden Open Variation Database (LOVD) (http://databases.lovd.nl/shared/genes/BRCA2); ClinVar (https://www.ncbi.nlm.nih.gov/clinvar/), BRCA exchange (https://brcaexchange.org/), and expert help were taken to tackle any confusion or problems that occurred during data extraction. In addition, to make the comparisons and notations uniform, all the information extracted on BRCA mutations incorporated Human Genome Variation Society (HGVS) (http://www.hgvs.org/rec.html) nomenclature recommendations, and all mutations are listed as either pathogenic/likely pathogenic, benign, a polymorphism, or a variant of unknown significance (VUS). Recurrent mutations are those reported twice or more in different articles. 21 Founder mutations were reported as they were designated in our included studies. Any disagreements were resolved with the help of a third reviewer (SY). Ethics approval was not required for this systematic review.
Results
We identified 493 studies from electronic database searches and four additional studies from manual searching of reference lists and related systematic reviews. After duplicate removal, we screened 170 articles by titles and abstracts. After screening, 57 full-text articles were retrieved and assessed against the predefined inclusion criteria, leaving 20 articles eligible to be included in the review. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) diagram detailing the identification and selection process is shown in Figure 1.

Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) diagram detailing the identification and selection process for relevant studies.
BRCA1 and BRCA2 mutations
This study screened a total of 1768 female BC patients and 573 BC families for exploring the BRCA1 mutations, and 1347 female BC patients and 573 BC families for BRCA2 mutations (Supplementary File 2).15,21–39
Bangladesh
One study (Akter et al. 2019) found the frequency of pathogenic mutations in BRCA1 and BRCA2 to be 4.65% (2/43) and 9.30% (4/43), respectively. 21
India
In 2002, Kumar et al. and Saxena et al. carried out a mutational analysis of BRCA1 and BRCA2.22,23 Only 3 (21.42%) out of 14 patients with a positive family history of breast or ovarian cancer had a BRCA1 mutation in the study by Kumar et al., while six sequence variants (two in BRCA1 and four in BRCA2) were found in a study by Saxena et al.
In 2004, Hedau et al. found that out of 100 sporadic BC cases analyzed, only exon 2 of both BRCA1 and BRCA2 in six patients (two (2%) in BRCA1 and four (4%) in BRCA2) showed alterations. 24 185delAG (c.68_69del), a frameshift mutation of BRCA1 in exon 2, was found to be the founder mutation among two patients.
Saxena et al. in 2006 found 18 sequence variants (9 distinct BRCA1 and 9 distinct BRCA2 variants). K1183R (c.3548_3549delinsGG), a variant leading to a lysine to arginine amino acid change, was the most common missense polymorphism reported among 16 patients. 25 However, the significance remains uncertain. In the same year, Syamala et al. revealed 13 distinct germline BRCA2 sequence variants in 94 patients. 26
Vaidyanathan et al. in 2009 found eight BRCA1 (20.5%) mutations with the majority (6/8; 75%) being frameshift mutations. They found only one BRCA2 mutation. 27 In the same year, Soumittra et al. found that 13 of the 71 samples analyzed had a deleterious mutation. Ten mutations were in BRCA1 and three were in BRCA2. 28
In 2012, Juwle et al. observed BRCA1 and BRCA2 mutations in 26 of 50 (52%) women diagnosed with early onset BC. 29 Mutations in BRCA1 were observed in 17 (34%) patients and mutations in BRCA2 in 12 (24%) patients. A missense mutation 5076G > A (c.4957G > A) was found among six women with BC.
Singh et al. (2015) identified 12 sequence variants of BRCA1 in the study group, including 10 frameshifts, 1 missense, and 1 nonsense mutation. 30
In 2016, Mannan et al. conducted a multi-gene sequencing analysis among 141 unrelated patients. 31 Fifty-one pathogenic cases (36.2%) included 19 novel mutations. 185delAG (c.68_69del), IVS17 + 1G > A (c.5074 + 1G > A), 3471C>T (c.3352C>T), and S1566fs (c.4837_4838delinsGCC) were recurrent mutations seen in BRCA1. Furthermore, 13/24 cases (54%) were found to have sequence variation; 69% (9 of 13 cases) were identified in BRCA1 and 31% (4 of 13 cases) in BRCA2 in a study by Darooei et al. in 2017 that performed whole genome sequencing. 32
In 2018, Mehta et al. and Shah et al. carried out mutational analyses.33,34 The frequency of mutations in BRCA1 and BRCA2 was 20 patients (15.9%) and 7 patients (5.6%), respectively, in a study by Mehta et al. Eleven cases with BRCA1 mutations (31.4%), 15 with BRCA2 mutations (42.9%), and 9 cases (25.7%) with both BRCA1 and BRCA2 mutations were found in a study by Shah et al. A missense mutation (87.8% of 35 cases) was the most common.
Pakistan
Liede et al. (2002) found that 6.7% of BC cases had either BRCA1 or BRCA2 mutations, with 65% of those cases accounting for BRCA1. IVS14-1G > A (c.4485-1G > A), 2080insA (c.1961dup), 4284delAG (c.4165_4166del), and 4184del4 (c.4065_4068del) in BRCA1 and 3337C>T (c.3109C>T) in BRCA2 were common mutations seen in most BC cases. 35
Rashid et al. (2006) conducted a BRCA mutation screening study in 39 cases of BC. A frameshift mutation, 185delAG (c.68_69del) in exon 2, was described as a founder mutation and found among two families, while a nonsense mutation, 4627 C>A (c.4508C>A), was described as a founder mutation among five families. 36
Moatter et al. (2011) studied 53 Pakistani BC patients consisting of 23 early onset BC cases and 30 BC patients with a family history. 37 They found only four mutations (two frameshifts, one missense, and a polymorphism mutation) in BRCA1.
In a 2019 study by Rashid et al., 133 deleterious mutations (24.7%) were found, including 110 in BRCA1 and 23 in BRCA2. 15 Frameshift mutation 3889delAG (c.3770_3771del) in exon 11 was the most common BRCA1 mutation found among 10 families, followed by nonsense mutation 5622C>T (c.5503C>T) among 9 families, exon 1-2 deletion among 7 families, frameshift mutation 804delT (c.685del) among 7 families, and intronic variant IVS14-1G > A (c.4485-1G > A) was found among 5 families. Frameshift mutation 5450delGTAA (c.5222_5225del) was found among four families and was the most common BRCA2 mutation among BC families. Large group rearrangements commonly found were deletions in exon 1-2, which were observed in seven BC families.
Sri Lanka
Silva et al. (2008) studied 130 Sri Lankan BC patients and found 19 sequence variants in BRCA1. 38 In 2017, Silva et al. found the prevalence of pathogenic and likely pathogenic variants of BRCA2 as 23% and 6.3%, respectively. 39 c.6728C>T in exon 11, a missense pathogenic BRCA2 mutation, was found among three Sri Lankan young BC patients.
Various intronic variants and VUS are described in some studies. The most common intronic variant was IVS14-1G > A (c.4485-1G > A), found among six Pakistani BC families in a study by Rashid et al. 15 and reported previously in the Breast Cancer Information Core Database (BIC).
Akter et al. showed a similar frequency in germline VUS as BRCA1 and BRCA2 mutations, 21 while Syamala et al. found all exonic or intronic sequence variations were VUS. 26 Four unclassified variants of unknown clinical significance [942G > A (c.823G > A), 3238G > A (c.3119G > A), 5002T>C (c.4883T>C), and 5076G > A (c.4957G > A)] were observed in nine patients in BRCA1, and nine missense mutations of unknown clinical significance in nine patients in BRCA2, were found in a study by Juwle et al. 29 Similarly, Darooie et al. also found four VUS, one in BRCA1 and three in BRCA2. 32 Most of the missense mutations as VUS were described in a study by Mehta et al. 33 , while Silva et al. found two intronic VUS (c.1910-74T>C and c.1910-51G>T) and another two VUS [(S775F (c.2324C>T) and P1702S (c.5104C>T)] in BRCA2. 39 VUS are outlined in detail in Tables S1 and S2.
The 185delAG (c.68_69del) mutation in BRCA1 and W31X (c.92G > A) mutation in BRCA2 were the most common recurrent mutations among Pakistani and Indian female BC patients.15,22,23,27,31,35(Tables S3 and S4)
Discussion
This review was done systematically to explore the spectrum of BRCA1 and BRCA2 gene mutations among female South Asian BC patients.
BRCA-associated BCs are seen at younger ages among Asian patients. A study in Korea showed approximately 50% of BC patients younger than 40 years of age have BRCA1/2 mutations. 40 This finding is consistent with most of the studies in our review.
The study on the Bangladeshi population found an enriched pathogenic BRCA2 mutation similar to that of Finnish, Chinese, and Cyprian BC patients.37,41,42 Studies on Indian and Pakistani people showed a higher frequency of BRCA1 mutations than BRCA2 mutations,15,23,24,27,32,35 as also observed in Saudi Arabia and most studies among white populations.43–45 In contrast, BRCA2 mutations have a higher incidence among Asian populations other than Indians and Pakistanis. In general, BRCA1 mutations are found more in other ethnicities.46,47 Similarly, a comprehensive analysis among Asian countries showed a higher frequency of BRCA1 mutations than that of BRCA2 mutations (622 vs 583). Yet, in some Asian countries, such as China, Hong Kong, Korea, and the Philippines, BRCA2 mutations had a higher frequency. 48
BRCA1/2 germline mutations with familial BC in Asian patients have a prevalence ranging from 8.0% to 31.8%. Early onset BC patients have a prevalence ranging from 2.8% to 21.4%. A recent Chinese cohort study found BRCA mutations in 9.1% of cases with at least one risk factor for hereditary BC, 3.5% of cases of sporadic patients, and 0.38% of healthy populations.46,49
Recurrent mutations are those repeatedly reported. The 185delAG (c.68_69del) BRCA1 mutation was first reported in Ashkenazi Jews, then in Chilean, Russian, and Israeli populations.50–53 185delAG (c.68_69del) is described as a founder mutation in Egyptian and Hungarian patients.54,55 In addition, this mutation in exon 2 is globally the second most frequent BRCA1 mutation, described in all Asian, American, African, and European populations. 56 The 3889delAG (c.3770_3771del) mutation in exon 12 of BRCA1 was previously reported among the Chinese and Malaysian populations.57,58 In exon 20, 5382insC (c.5266dupC) was reported first in the Danish population. It is common in European countries, is the most frequent BRCA1 mutation in the world, and is described as a founder mutation in Russian and Turkish populations.49,56 This mutation was seen in only one family in the study by Rashid et al. 2019. In European countries, 5382insC (c.5266dup) is the most important and prevalent BRCA1 mutation even though Asian and American BC patients rarely have it. 59 The three most well-characterized founder mutations in Ashkenazi Jews are two in BRCA1 (185delAG (c.68_69del) and 5382insC (c.5266dup)) and one in BRCA2 (6174delT (c.5946delT)).60–62 Screening for these three founder mutations alone is now part of routine clinical practice for Ashkenazi Jewish individuals. 63 The 5382insC (c.5266dup) mutation in BRCA1 exon 20 is the second most frequently reported mutation in the BIC database, being prevalent in central and eastern Europe. 63 However, the mutation pattern among BC patients has varied widely according to the countries. Additionally, in the BC families who were genotyped from Boston, Massachusetts, USA, 6697delTC (c.6468_6469del) was the most frequent mutation accompanying the three major Ashkenazi Jew insertions/deletions. 56
In exon 11 and exon 14 of BRCA1, 3889delAG (c.3770_3771del) and 5622C>T (c.5503C>T), respectively, which were found among the Pakistani population, were previously described among Turkish and Filipino individuals.15,64,65 A missense mutation 5076G > A (c.4957G > A) in exon 16 with amino acid change M1652I (c.4956G > A) was found among early onset BC patients of Indian ethnicity. This was reported earlier in a study by Thompson et. al. 66 The Y130X (c.390C>A) BRCA1 mutation, found mainly among Japanese and Korean patients, was the second most reported mutation in Asia. BRCA2 mutations that are commonly found mainly in Korean and Chinese patients were 7708C>T (c.7480C>T) (53 cases; 11 BIC entries), K467X (c.1399A>T) (29 cases; two BIC entries), and 3972del4 (c.3744_3747delTGAG) (26 cases; eight BIC entries). 48 In a study by Kwong et al., two mutations were referred to as recurrent mutations. The 589delCT (c.470_471delCT) mutation of BRCA1 was observed among Chinese and Korean patients in Hong Kong, Malaysia, and the USA, and described previously in Japanese and Pakistani patients. BRCA1 1100delAT (c.981_982delAT) is described in Chinese patients. These recurrent mutations were 20.6% of all BRCA1 mutations. 67 These mutations are also described in our study, but have low frequency in South Asian regions. As seen in our study and in accordance with BIC, most of the BC-causing pathogenic mutations in BRCA1 and BRCA2 are often nonsense, frameshift, and splicing mutations that lead to the production of a truncated protein. 68
The contribution of large genomic rearrangements (LGRs) in Asian high-risk BC patients has been described in a few studies. LGRs accounted for 6.3% of all mutations in BRCA1/2 in a Malaysian cohort. 69 A study from southern China found LGRs in 0.7% (4/555) of high-risk breast or ovarian cancer patients in BRCA genes and accounted for 5.8% of overall BRCA1/2 mutations in the study cohort. 70 Our review also included a few studies accounting for LGRs. LGR mutation del exon 1-2, found among seven BC families, was previously reported in the Omani population. 71 An increasing number of reports on polymorphisms, rare sequence variants, and missense mutations as likely pathogenic, along with LGRs, were found among different ethnic groups. 48
K1183R (c.3548_3549delinsGG), a common missense mutation in BRCA1 and a VUS found in our study, was previously reported by Thompson et al. 66 The frequency of VUS differs according to the ancestry of individuals and testing laboratories. The African American population showed the highest rate of VUS (21%), followed by European (15%) and Americans with European ancestry (5%–6%).72,73 VUS accounting for 40% of total variants are not distinctly established and classified, even when next-generation sequencing (NGS) was used. The use of VUS in a clinical setting is quite challenging, and the clinical and genetic significance of VUS has not been clearly delineated in the literature to date. 74
The BRCA1 and BRCA2 gene mutation frequencies are not very different between different countries within South Asia. The similarity in race and ethnicity among these countries is one of many possible reasons for the similar mutation pattern, as shown in Tables S1 and S2.
With recent advancements in genetic testing, NGS methodology can be used to analyze the entire human genome in a single day. This can more accurately identify the various mutations that predispose an individual to increased risks for developing various cancers. Though much progress has been made after the development of NGS, many BRCA1 and BRCA2 mutations and their roles in pathogenicity are yet to be elucidated. Thus, a better understanding of mutations in these genes with consideration of different ethnic populations is crucial. 48 Moreover, genetic profiles from clinical assessments provide information on the degree of admixture and susceptibility to genetic disease.75,76 For optimizing efficient strategies for genetic testing for BRCA1 and BRCA2 mutations, information on the high frequency of these mutations among Asian populations with those residing in various parts of the world is of great need. 48
A lack of uniform nomenclature, differences in diagnostic interventions used, non-specific terminologies, and unavailability of complete datasets caused variation in reporting. Thus, the authors' consensus and expert help were considered while entering data. Similarly, the differences in ascertainment, like age at diagnosis or family history, led to the varied frequencies of pathogenic variants. Therefore, the recurrent and common mutations may not be a true reflection of South Asian BC patients. Furthermore, studies were included from only four South Asian countries. Additionally, the majority of the BC cases were sporadic with an estimated frequency ranging from 90% to 95%. The remaining cases (5%–10%) included familial BCs. 77 However, the majority of included studies provided no information on whether the mutation was sporadic or germline/familial.
The knowledge of the local mutation spectrum pattern can help the concerned authorities and researchers make possible plans for future epidemiological studies in the larger population. It can also help to conduct pilot screening studies for genetic testing and mutational analysis among women and different ethnic groups within the country. In addition, the establishment of an Asian registry of BRCA1/BRCA2 mutation carriers seems paramount to allow more organized research work to be done on this population. Having an understanding of the most recurrent BRCA1 and BRCA2 mutations of our ethnicities or population can help improve and speed up the diagnosis, treatment, and follow-up processes for BC patients. Elucidating the founder effect of BRCA1/2 genes can significantly impact the management of hereditary cancer families on a national and international healthcare system level, making genetic testing more affordable and cost-effective.
Conclusions
The South Asian population has a wide variety of genetic mutations of BRCA1 and BRCA2, differing according to the countries and ethnicities. An improved knowledge of the BRCA1/BRCA2 mutation spectrum can help clinicians deliver proper genetic counseling and cancer management to women diagnosed with or at high risk of developing BC.
Preprints available
Doi: 1) 10.21203/rs.3.rs-290397/v1(https://www.researchsquare.com/article/rs-290397/v1)
2) 10.21203/rs.3.rs-290397/v1 (https://europepmc.org/article/ppr/ppr356734)
Supplemental Material
sj-pdf-1-imr-10.1177_03000605211070757 - Supplemental material for BRCA1/BRCA2 mutation spectrum analysis in South Asia: a systematic review
Supplemental material, sj-pdf-1-imr-10.1177_03000605211070757 for BRCA1/BRCA2 mutation spectrum analysis in South Asia: a systematic review by Sanjeev Kharel, Suraj Shrestha, Siddhartha Yadav, Prafulla Shakya, Sujita Baidya and Suzita Hirachan in Journal of International Medical Research
Supplemental Material
sj-pdf-2-imr-10.1177_03000605211070757 - Supplemental material for BRCA1/BRCA2 mutation spectrum analysis in South Asia: a systematic review
Supplemental material, sj-pdf-2-imr-10.1177_03000605211070757 for BRCA1/BRCA2 mutation spectrum analysis in South Asia: a systematic review by Sanjeev Kharel, Suraj Shrestha, Siddhartha Yadav, Prafulla Shakya, Sujita Baidya and Suzita Hirachan in Journal of International Medical Research
Supplemental Material
sj-pdf-3-imr-10.1177_03000605211070757 - Supplemental material for BRCA1/BRCA2 mutation spectrum analysis in South Asia: a systematic review
Supplemental material, sj-pdf-3-imr-10.1177_03000605211070757 for BRCA1/BRCA2 mutation spectrum analysis in South Asia: a systematic review by Sanjeev Kharel, Suraj Shrestha, Siddhartha Yadav, Prafulla Shakya, Sujita Baidya and Suzita Hirachan in Journal of International Medical Research
Footnotes
Author contributions
SK and SS were involved in the conceptualization of the study, along with designing the study search strategy, reviewing study abstracts, extracting data from full-text articles, and drafting the initial manuscript. SY, PS, SB, and SH were involved in editing and revising the manuscript. All the authors read and approved the final version of the manuscript.
Declaration of conflicting interest
The authors declare that there is no conflict of interest.
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Supplemental material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
