Abstract
Background
Increasing use of BRCA1/2 testing for tailoring cancer treatment and extension of testing to tumour tissue for somatic mutation is moving BRCA1/2 mutation screening from a primarily prevention arena delivered by specialist genetic services into mainstream oncology practice. A considerable number of gene tests will identify rare variants where clinical significance cannot be inferred from sequence information alone. The proportion of variants of uncertain clinical significance (VUS) is likely to grow with lower thresholds for testing and laboratory providers with less experience of BRCA. Most VUS will not be associated with a high risk of cancer but a misinterpreted VUS has the potential to lead to mismanagement of both the patient and their relatives.Design
Members of the Clinical Working Group of ENIGMA (Evidence-based Network for the Interpretation of Germline Mutant Alleles) global consortium (www.enigmaconsortium.org) observed wide variation in practices in reporting, disclosure and clinical management of patients with a VUS. Examples from current clinical practice are presented and discussed to illustrate potential pitfalls, explore factors contributing to misinterpretation, and propose approaches to improving clarity.Results and conclusion
Clinicians, patients and their relatives would all benefit from an improved level of genetic literacy. Genetic laboratories working with clinical geneticists need to agree on a clinically clear and uniform format for reporting BRCA test results to non-geneticists. An international consortium of experts, collecting and integrating all available lines of evidence and classifying variants according to an internationally recognized system, will facilitate reclassification of variants for clinical use.Free full text
BRCA1 and BRCA2 genetic testing—pitfalls and recommendations for managing variants of uncertain clinical significance
Abstract
Background
Increasing use of BRCA1/2 testing for tailoring cancer treatment and extension of testing to tumour tissue for somatic mutation is moving BRCA1/2 mutation screening from a primarily prevention arena delivered by specialist genetic services into mainstream oncology practice. A considerable number of gene tests will identify rare variants where clinical significance cannot be inferred from sequence information alone. The proportion of variants of uncertain clinical significance (VUS) is likely to grow with lower thresholds for testing and laboratory providers with less experience of BRCA. Most VUS will not be associated with a high risk of cancer but a misinterpreted VUS has the potential to lead to mismanagement of both the patient and their relatives.
Design
Members of the Clinical Working Group of ENIGMA (Evidence-based Network for the Interpretation of Germline Mutant Alleles) global consortium (www.enigmaconsortium.org) observed wide variation in practices in reporting, disclosure and clinical management of patients with a VUS. Examples from current clinical practice are presented and discussed to illustrate potential pitfalls, explore factors contributing to misinterpretation, and propose approaches to improving clarity.
Results and conclusion
Clinicians, patients and their relatives would all benefit from an improved level of genetic literacy. Genetic laboratories working with clinical geneticists need to agree on a clinically clear and uniform format for reporting BRCA test results to non-geneticists. An international consortium of experts, collecting and integrating all available lines of evidence and classifying variants according to an internationally recognized system, will facilitate reclassification of variants for clinical use.
introduction
Germline inactivating variants in the tumour suppressor genes BRCA1 and BRCA2 confer high lifetime risks of breast cancer, ovarian cancer and less frequently also other cancers [1]. These pathogenic variants are conventionally called ‘mutations’ or ‘deleterious variants’ in BRCA genetic testing parlance and the term pathogenic variant is used here for the sake of precision. Pathogenic variants in either gene confer a high lifetime risk of developing ovarian or (another) primary breast cancer in female carriers but they explain only ~20% of familial breast cancer [2]. The more cancers, the younger the onset and the admixture of ovarian with breast cancer among relatives all increase the chance that a familial cluster is due to a BRCA gene mutation but, nonetheless, most familial clusters of breast cancer are not due to an inherited mutation in BRCA1 or BRCA2. In addition to familial clusters, BRCA mutations account for over 10% of patients with early onset triple-negative breast cancer and over 10% of women with non-mucinous ovarian cancer unselected for family history [3, 4]. Female carriers with pathogenic variants can make informed decisions about prophylactic surgery or intensified screening programmes. High-profile media coverage increases patient expectations from genetic testing [5, 6]. The indications for germline genetic testing for BRCA1 and BRCA2 are further increasing to include directing cancer chemotherapy, novel targeted treatments and informing choices about the extent of therapeutic surgery [7–9]. Germline genetic testing for BRCA1 and BRCA2 for cancer risk prediction and management is routinely delivered by clinical genetics professionals but increasing demand is overwhelming the current delivery model with insufficient capacity among trained geneticists and genetic counsellor. Safe integration of genetic counselling and testing to identify Hereditary Breast and Ovarian Cancer in mainstream oncology is an ongoing challenge. Challenges exist at many levels but key are genetic literacy and genetic test outcome interpretation and reporting.
Variability between individuals' genetic code is common within the general population and between individuals of different ethnic background, and this intrinsic variability can lead to difficulties in interpreting some types of sequence change (see Figure Figure11 for more information). Variants of uncertain clinical significance (VUS) represent a particular challenge since the clinical significance cannot be inferred from sequence information alone. Misinterpretation of VUS can lead to real clinical harms for both patients and families [10–12]. Furthermore, terms used in genetic test clinical reports vary (see Table Table1).1). Up to 20% of BRCA1/2 tests will report genetic VUS but, in a well-characterized ethnic population, the proportion may drop to 5% or less [13–15]. The Evidence-based Network for the Interpretation of Germline Mutant Alleles (ENIGMA) consortium has received over 6000 submissions of unique VUS, identified in over 13 000 families from over 17 countries [16]. Laboratories following generic reporting rules and with limited BRCA-specific experience may result in reports with more categoric conclusions about variants in BRCA genes than is supported when multiple lines of evidence are taken into account [17]. The percentage of gene tests resulting in a VUS is expected to increase when the extent of sequencing increases to include untranslated and deeper intronic regions and when tumour testing is offered. We restrict examples in this paper to the BRCA genes tested in blood samples.
Table 1.
Types of sequence variants reported in BRCA mutation detection tests according to risk relevant for clinical management
| Clinical risk | Descriptors observed in clinical reports | Interpretation |
|---|---|---|
| High | Functionally deleterious mutation Pathogenic mutation High-risk mutation Deleterious variant | Variants that result in a high lifetime risk of breast and ovarian cancer. Deleterious variants disrupt normal protein function. They include nonsense changes, out-of-frame insertions or deletions, large gene rearrangements or splicing variants altering the canonical splicing acceptor and donor sites disrupting regulatory regions, as well as some missense changes. Supportive evidence from multiple sources may be required to call a variant deleterious where predictions of functional consequences are uncleara. |
| Uncertain | Missense mutation Rare variant Variant of uncertain clinical significance (VUS, VOUS) Uncertain variant Variant of uncertain Variant of unknown clinical significance | Variants that differ from the published reference DNA sequence (RefSeq BRCA1: U14680; BRCA2: NM00059) and are not classifiable as either deleterious or as neutral based on available evidence. The association with ‘clinical phenotype’ at this time is ‘unknown’ and cannot be used to inform clinical decisions. |
| None | Frequent mutation Common variant Polymorphism Variants of low clinical significance Variants of no clinical significance Neutral variants Benign variants | Changes in the DNA sequence that do not disrupt the normal function of the encoded protein and are not associated with a clinically important increased risk of disease. If present in the population at a frequency of >1%, these may be referred to as polymorphisms. |
aFor example assays showing partial loss of normal transcript from aberrant splicing or partially reduced function from an amino acid substitution may require additional evidence to classify as class 4 or 5.
Genetic variability information.
Members of the ENIGMA consortium's Clinical Working Group have collated real case scenarios from clinical practice to illustrate the pitfalls that can arise after a VUS test result and suggest some strategies to mitigate this risk. Clinicians who are requesting BRCA tests need to consider these issues when using BRCA test results in the management of patients and their families.
the VUS report: why are VUS difficult to classify?
pitfalls associated with variability in the information content of genetic test reports
Based on the experience of representatives from 17 countries, it is clear that there is currently no internationally accepted standard for BRCA testing reporting (Table (Table1),1), and no agreed consistent classification system: some laboratories report variants without interpretation, some use a narrative approach and some use locally developed guidelines or published schemes [18, 19]. For some types of BRCA gene variants, additional evidence may be essential before a variant can be clearly classified. Different lines of evidence may appear to conflict, so an integrated estimation of probability taking all available evidence into account is essential to reach a final classification for many variants [20, 21].
multiple lines of evidence may be required to establish pathogenicity
Software packages are available to help interpret genetic variants, each with strengths and weaknesses (Table (Table2).2). Understanding of how the multiple functions of the BRCA1 and BRCA2 proteins relate to cancer predisposition is limited and few validated functional assays are available [22]. The interpretation of functional studies is often technically complex, and results may not be calibrated against clinical parameters in order to give measures of sensitivity and specificity. A major breakthrough in the field of BRCA variant classification has been the development of a multifactorial likelihood classification model [20, 23]. It is anticipated that functional data from BRCA1 and BRCA2 will soon be incorporated in this model [22, 24, 25].
Table 2.
Examples of software tools available for clinical variant evaluation online
| Resource | Description | Website |
|---|---|---|
| Polyphen2 | POLYmorphism PHENotypes Polyphen2 is freely available, web-based program from Harvard University that predicts whether an amino acid substitution affects protein structure and function. | http://genetics.bwh.harvard.edu/pph2/ |
| SIFT | Sorting Intolerant From Tolerant SIFT software is freely available and predicts whether an amino acid substitution affects protein function. | http://sift.jcvi.org/ |
| Align-GVGD | Align Grantham Variation and Grantham Deviation. Align-GVGD is a freely available, web-based program from International Agency for Cancer Research (IARC) that combines the known chemical nature of amino acids and the alignments of DNA sequences across species to predict the likely pathogenicity of missense substitutions. | http://agvgd.iarc.fr/about.php |
| HCI database of prior probabilities | The Huntsman Cancer Institute Database of Prior Probabilities of Pathogenicity for single-nucleotide substitutions provides the prior probability of pathogenicity estimate which is the starting point for the multifactorial likelihood estimate for a novel SNP identified through genetic testing for a BRCA gene. | http://priors.hci.utah.edu/PRIORS/ |
| GeneSplicer | GeneSplicer is a freely available flexible system for detecting splice sites in the genomic DNA of various eukaryotes. | http://www.ncbi.nlm.nih.gov/pmc/articles/PMC29713/ |
| Human Splicing Finder | This tool is freely available and aimed at studying pre-mRNA splicing and identifying variants that may disrupt normal splicing. | http://www.umd.be/HSF3/ |
| Commercial software, e.g. Alamut | Alamut is commercially available software that provides a single interface to bring together multiple data sources including many of those listed above in a simplified and streamlined tool for rapid assembly of data about any one variant aimed at busy diagnostic laboratories. | http://www.interactivebiosoftware.com/software/alamut/overview |
| Entrez-Pubmed Google Scholar | Web-based search engines to look for research publications including a specific mutation. | http://www.ncbi.nlm.nih.gov/pubmed http://scholar.google.com |
| UMD-BRCA1 UMD-BRCA2 | The UMD-BRCA1/BRCA2 databases have been set up in a joined national effort through the network of the 16 French diagnostic laboratories. A classification is proposed based on the classical parameters including splicing algorithms. A strong effort is done to perform co-segregation studies. | http://www.umd.be/BRCA1/ http://www.umd.be/BRCA2/ |
The multifactorial likelihood method uses a number of different independent features in order to establish a combined likelihood estimate that a BRCA variant has the characteristics of known pathogenic variants [15]. The model currently combines the prior probability of variant pathogenicity (based on evolutionary conservation and amino acid physicochemical properties) with additional estimates of pathogenicity from likelihoods based on clinical information, including variant co-occurrence with a known pathogenic mutation in the same gene, tracking of the variant with cancer-affected family members (segregation) and BRCA tumour features. The concept underlying multiparametric methods is that empiric probabilities that a person ‘does’ have or ‘does not’ have a pathogenic BRCA variant can be established for each available line of evidence [20]. Multiple lines of evidence can be factored together as they become available in an iterative manner to produce an increasingly accurate probability of pathogenicity estimates for an individual variant. These estimates can be used to classify variants into clinically relevant strata (Table (Table3)3) [18].
Table 3.
5-tier classification BRCA1/2 VUS modified from ref. [15]
| Class | Description | Probability of being pathogenic | Clinical predictive testing of at risk relatives | Management recommendations if at-risk relative has the variant | Research testing of family members |
|---|---|---|---|---|---|
| 5 | Definitely pathogenic | >0.99 | Yes | Full high-risk guidelines | Not indicated |
| 4 | Likely pathogenic | 0.95–0.99 | Yes | Full high-risk guidelines | May be helpful to further classify variant |
| 3 | Uncertain | 0.05–0.949 | No | Presence of variant is irrelevant to risk assessment, manage risk based on family history only | May be helpful to further classify variant |
| 2 | Likely not pathogenic or of no clinical significance | 0.001–0.049 | No | Manage risk based on family history only | May be helpful to further classify variant |
| 1 | Not pathogenic or of no clinical significance | <0.001 | No | Manage risk based on family history only | Not indicated |
several classification systems have been proposed and there is no international consensus on which to adopt
A universal system for classification of variants common across all countries would facilitate education among new users of genetic tests and minimize the risk of misinterpretation. Unfortunately, multiple systems have been proposed and are currently in use with many reporting laboratories not offering a clinically relevant classification. A World Health Organization-funded expert workshop at the International Agency for Research on Cancer (IARC) in 2008 recommended that the score from a multifactorial likelihood model is used to categorize high-risk cancer gene variants [26] based on the multifactorial likelihood estimates of variant pathogenicity [18]. This is the only published classification system that links clinical recommendations to each class (Table (Table3).3). The American College of Medical Genetics are recommending ClinVar as a common repository for genetic variants to help standardise reporting in the United States (Table (Table44).
Table 4.
Commonly used online database resources that provide some interpretation of BRCA sequence variants although
| Resource | Description | Website |
|---|---|---|
| IARC/LOVD [Leiden Open (source) Variation Database] | Separate BRCA1 and BRCA2 variant databases curated by experts at the University of Leiden and IARC. Includes only those VUS with literature references and records the published literature associated which each variant. Each variant is given an IARC class [15] and links to published source data. Submissions of variants are from registered submitters largely from the research community. Curation of submissions for each gene is undertaken by a named expert curator on a voluntary basis. | http://brca.iarc.fr/LOVD/home.php soon to be replaced by http://hci-exlovd.hci.utah.edu/home.php |
| The Breast Cancer Information Core (BIC) | A database that acts as a central repository for BRCA1 and BRCA2 variants (deleterious, neutral or VUS) deposited by submitters from research and clinical sites internationally. Recently, a central curation process working to classify all variants according to the IARC 5-tier classification [15] scheme has been introduced. | https://research.nhgri.nih.gov/projects/bic |
| ClinVar | ClinVar is a freely accessible, public archive of reports of the relationships between human variations and phenotypes presented with supporting evidence and an indication of likely clinical significance. Submissions are from research and some diagnostic laboratories and submitters include a classification for submitted variants, submissions are not currently curated. | http://www.ncbi.nlm.nih.gov/clinvar/ |
| Human Variome Project | The Human Variome Project, under the auspices of UNESCO, has been created as a unified reporting portal and lists four separate databases for BRCA1 and four for BRCA2. As yet it offers no formal curation or attempt to classify variants. | http://www.humanvariomeproject.org |
databases reporting genetic variants
A number of web-based resources catalogue variants reported in the BRCA1 and 2 genes (Table (Table4).4). For all variant databases, ongoing curation of deposited information, both at the time of deposition and reclassification, is a significant challenge. Some databases are actively curated while others rely heavily on the classification by the submitting laboratory with varying levels of supporting data provided. Also important is the use of standardized HGVS nomenclature (http://www.hgvs.org/) allowing unambiguous comparison of all data on the same variant in the database and across the literature. Apart from occasional national consortia, reporting of variants is not mandated so databases cannot usually be used to derive population frequencies of variants to aid classification (e.g. laboratories tend to under-report variants that have been found several times) and many do not catalogue the supporting evidence used to determine pathogenicity. A new attempt to collate BRCA VUS for the purpose of classification is the PROMPT registry (www.promptstudy.org). Patients receiving a BRCA VUS result are informed about the registry directly by the collaborating testing laboratory (including Myriad and Ambry Genetics) and/or their clinician. The patient can submit a range of medical information, genetic test report and tumour pathology reports and opt to participate in a variety of variant-specific research studies.
Case 1: Two families were identified in separate countries with the same missense variant in BRCA1 designated c.5212G > A, NM_007294.3 (p.Gly1738Arg). A number of women in both families were seeking risk-reducing surgery because of a strong family history of early age of onset of breast cancer in their families. The variant was a class 3 (uncertain) variant (Table (Table3)3) but it occurred at a highly conserved residue, and it was recognized that more evidence may successfully confirm it was pathogenic. Predictive testing could then be offered to at-risk family members to facilitate their preventive choices. Each family provided insufficient power for an informative segregation analysis but collaboration across centres in two countries with the same variant, allowed a segregation analysis to be completed and the VUS could then be re-classified as class 5 (definitely pathogenic) [27]. It took a fortuitous academic collaboration and over 3 years to re-classify this VUS. Unless the phenotypes and genotypes are deposited in a freely accessible database, then such a variant will remain class 3. The ENIGMA consortium [16] provides a mechanism to rapidly link clinical teams looking after families with the same VUS to pool evidence. Since BRCA germline mutations explain only a minority of all cases presenting with breast or ovarian cancer, two unrelated patients with these relatively common phenotypes and the same rare BRCA1 variant provide insufficient evidence of pathogenicity. In this scenario, additional evidence from segregation analyses is essential to allow a multifactorial estimation of the likelihood of pathogenicity before confidently using the variant in clinical predictive testing.
pitfalls derived from frequency and population of origin of the VUS
Rare variants (allele frequency <0.01) are usually not classifiable by an individual laboratory due to paucity of information and lack of statistical power. When the ethnicity of the patient being tested differs from the patient groups where most testing has been done, a sequence variant may have little publically available data and be rare in that laboratory so is more likely to be considered a VUS.
Case 2: A 35-year-old woman of African ancestry developed two primary breast tumours presenting at age 25 and 33. Both tumours were estrogen receptor negative (ER-), progesterone receptor negative (PR-) and human epidermal growth factor receptor 2 negative (HER2-). She had no family history of cancer, but testing was initiated on the basis of her personal cancer history. She was found to have two VUS in BRCA1; variant 1, c.5154G > T, NM_007294.3 (p.Trp1718Cys) and variant 2 in the 3′ UTR region, c.*36C > G, NM_007294.3. DNA samples from relatives indicated that both variants are likely located on the same allele. Variant 2 is frequent (6%–11%) in the African-American population [28] and could be assigned to class 1 (non-pathogenic). Variant 1 involves amino acid substitution at a highly conserved residue predicted in silico to have a functional impact but could not be classified unambiguously due to insufficient data. The clinical dilemma for this patient thus becomes whether the available information about variant 1 is sufficient to predict a high risk of ovarian cancer and recommend risk-reducing bilateral salpingo-oophorectomy (BSO).
A literature search identified two reports of variant 1. The first report was of two families from Asturias, Northern Spain [29]. Contact with the authors clarified the following details, some directly applicable for variant interpretation: Asturian proband 1 was diagnosed with bilateral invasive ductal breast cancer at age 28 (grade III, ER−) and 30 (grade II, ER−, PR−, HER2–); Asturian proband 2 developed breast cancer at age 40 (ER and grade unknown), and probably also ovarian cancer at age 63. The second report of the variant was in an African-American patient and family members where the variant was present in 4/4 women with cancer and absent in 3/3 at-risk women without cancer. Pathogenicity was supported by the functional assay showing loss of transcriptional activity [30].
Using multifactorial likelihood analysis [23] combining all currently available evidence: a prior probability of 0.81 based on Align-GVGD prediction [31], and likelihood ratios based on additional information gained about breast tumour pathology [32], the posterior probability of pathogenicity for the variant is 0.98 which places it in class 4. Thus, sufficient evidence is now available to estimate a substantially raised ovarian cancer risk and permit a clinically sound recommendation of BSO to reduce ovarian cancer risk.
why is genetic counselling around VUS complex?
Although it is expected that most class 3 variants will represent non-pathogenic variants, it is critical that we improve and accelerate clinical annotation and classification.
limited genetic literacy among clinical professionals
Disclosure of a VUS result can test the skills of even highly experienced genetic counselling teams. There can be considerable variation in experience with molecular genetics and with clinical BRCA gene testing leading to diverse clinical management recommendations. This situation is compounded by complex and variable presentation of diagnostic reports from laboratories and limited genetic literacy among treating clinicians, and exacerbated further by the patient's perception, prior experiences and beliefs.
Case 3: A BRCA2 missense variant p.Thr3033Ile (c.9098C > T, NM_000059.3) was identified in a 53-year-old woman with ovarian cancer and no family history of breast or ovarian cancer. A class 3 report was issued as little supporting evidence was available. The managing clinician treated the variant as clinically significant and offered genetic testing to the patient's daughter and sister. The patient's sister did not carry the variant, but the daughter did. The daughter was given the same cancer risk management advice as a pathogenic variant carrier including risk-reducing surgery. Additional information from a BRCA2 functional assay was subsequently used to reclassify this variant to class 2 (likely of no clinical significance) [33]. Cancer risk estimates for the daughter and sister should have been based on family history alone. Since there was no family history, testing in relatives could not even provide useful additional information about segregation. It is worth noting here that the majority of breast and ovarian cancer patients who will be offered germline BRCA testing to guide cancer treatment will not have a family history of cancer amenable to segregation studies.
This case illustrates the need to ensure improved genetic literacy among non-genetics professionals ordering DNA tests, particularly the uncertainty around VUS results, the implications of testing in other family members and the need for a clear pathway for referral to colleagues experienced in the interpretation and follow-up of variant results [34]. Many patients will readily assume that any reported variant must be pathogenic unless the potential for a VUS has been clearly flagged as part of the pre-test discussion.
perceptions of cancer risks associated with VUS
A genetic test for a disease like breast cancer that uncovers a rare variant may lead to both patients and clinicians to think, ‘what is the chance this rare variant has nothing to do with this clinical presentation?’ and convince themselves it cannot be coincidental. The decision to undertake preventive surgery is complex and patients are likely to have had personal and family experiences which may have strongly contributed to their risk management decision. Class 3 VUS reports discussed with counselees are too frequently inaccurately perceived typically leading to overestimation of cancer risks, adverse psychological outcomes and more radical medical decisions [35, 36]. Furthermore, ‘almost the same large number of counselees with an unclassified variant decided to have preventive surgery as pathogenic mutation carriers’ [37].
Case 4: A BRCA1 VUS was identified some years previously in a woman who developed breast cancer at the age of 45 in the setting of a strong family history of early onset breast and ovarian cancer. The family understood that the variant must be the answer to the strong family history despite the inconclusive evidence. Several family members came forward for predictive genetic testing which they incorrectly believed was available. Genetic testing was repeated some years later and a clearly pathogenic large exonic deletion in BRCA1 was then identified on the opposite allele confirming irrefutably that the original VUS could not be pathogenic and demonstrating that new or optimized mutation screening approaches will uncover missed mutations in some families.
This case illustrates the importance of understanding a ‘non-informative’ or inconclusive test aimed at detecting a genetic predisposition to cancer. It is important to be aware that mutation testing is <100% sensitive and a negative BRCA test result does not exclude an underlying hereditary cause. BRCA1 or BRCA2 gene mutations only account for an estimated 20%–30% of familial clustering of breast cancer, so cancer patients negative for a BRCA test, but with a strong family history, should be referred for specialist assessment and advice about risk management in the cancer genetics clinic.
The testing clinician has a responsibility to ensure that they have received appropriate training about the genetic testing process and understand the possible outcomes and have appropriate expert support from a genetic specialist. A helpful starting point is to be able to estimate the probability of a deleterious BRCA mutation given the family history and tumour characteristic [38]. Testing clinicians should be able to explain the likely outcomes of the test to their patients; in patients with a low a priori probability of finding a pathogenic variant, the most likely outcome from testing would be no pathogenic variant found, the second most likely outcome would be a VUS, and the least likely outcome a pathogenic variant. It is good practice before embarking on a BRCA gene test to discuss with the patient the likelihood of these possible outcomes. The test result must be interpreted in the context of the family history and clear guidelines should be agreed for referral of cases to a specialist genetic clinic (Figure (Figure2).2). Finally, the experience gathered with VUS in BRCA1/2 is relevant to developing ethical norms and policy issues, including duty to re-contact patients, as cancer gene panel, exome and genome sequencing become increasingly commonplace [38].
Suggested model of a clinical pathway for breast and ovarian cancer patients. Initial genetic testing through an oncologist with basic training, expertise routinely contributed by the clinical genetics consultant. This model works where there is a close working relationship between oncologist and geneticist through a regular multidisciplinary team meeting.
recommendations
We believe that some essential elements are necessary for the optimal clinical use and interpretation of VUS in mainstream medical practice.
variant reporting and classification
An internationally accepted terminology and a clinically relevant classification to report and discuss BRCA test results.
A framework for clinicians, clinical scientists and research groups to work together towards classification of VUS.
Reporting of sequence variants to a single, transparently and expertly curated database providing clinically relevant classification for each reported BRCA variant based on clearly stated lines of evidence and freely available to all providers of genetic testing. To submit variants to ENIGMA, go to contacts page http://www.enigmaconsortium.org/
risk communication
Improve the genetics literacy of medical providers by structured training and integrating genomics into undergraduate, basic and specialist medical training curriculae.
Access and clear referral guidelines to specialist clinical genetics services for patients with a pathogenic variant, a VUS or a strong family history and no detectable genetic cause.
A close working relationship between genetic diagnostic laboratories, genetic specialists and cancer clinicians delivering an integrated care pathway for patients and their families.
data review
A standard operating procedure within reporting laboratories to review variant classification each time new evidence emerges.
A clearly agreed process for reporting updated classifications (laboratories) and re-contacting patients (clinicians) if a VUS becomes classified into clinically actionable or definitively non-pathogenic.
concluding remarks
VUS identified in BRCA genes represent a major clinical challenge. Individuals with significant family history, a pathogenic variant or a VUS should be referred to a genetic specialist service. Patients with no pathogenic variant or a VUS should be managed based on the family history only. Standardized reporting and better genetic literacy must be implemented to safely introduce genetics into mainstream oncology. In conclusion, concerted action between the clinical and research communities is the best approach to optimally managing BRCA variants for maximum patient benefit. The international ENIGMA consortium is focussed on improving the interpretation of BRCA VUS and incorporates both clinical and research expertise. Through a collaborative approach and a global and unified source of data and variant classification, important advances in this complex field will continue to benefit patients.
discloure
The authors have declared no conflicts of interest.
appendix
Name and affiliations of the ENIGMA CWG contributors: (excluding those named in the authors' list). R. Driessen: Familial Cancer Centre, Peter MacCallum Cancer Centre, The University of Melbourne, Melbourne, Australia. N. M. Lindor: Department of Health Sciences Research, Mayo Clinic, Scottsdale, USA. M. J. Blok: Department of Clinical Genetics, Maastricht University Medical Centre, Maastricht, the Netherlands. P. Moller: Research Group Inherited Cancer, Department of Medical Genetics, University Hospital, Oslo, Norway. M. de la Hoya: Molecular Oncology Laboratory, Hospital Clínico San Carlos (IdISSC), Madrid, Spain. T. Pal: Cancer Epidemiology Program, H. Lee Moffitt Cancer Center, Tampa, USA. S. Domchek: Basser Research Center and Abramson Cancer Center University of Pennsylvania Philadelphia, USA. K. Nathanson: Basser Research Center and Abramson Cancer Center University of Pennsylvania Philadelphia, USA. C. Van Asperen: Department of Clinical Genetics, Leiden University Medical Center, Leiden, the Netherlands. O. Diez: Oncogenetics Group, University Hospital Vall d'Hebron; Vall d'Hebron Institute of Oncology (VHIO), Barcelona, Spain. K. Rheim: Center for Hereditary Breast and Ovarian Cancer, Center for Integrated Oncology (CIO) and Center for Molecular Medicine Cologne (CMMC), University Hospital Cologne, Germany. D. Stoppa-Lyonnet: Genetics Service, Curie Institute, University of Paris Descartes, Paris, France. M. Parsons: Molecular Cancer Epidemiology Laboratory, Genetics and Computational Biology Division, QIMR Berghofer Medical Research Institute, BNE, Herston, QLD, Australia. D. Goldgar: Department of Dermatology, Huntsman Cancer Institute, University of Utah School of Medicine; Salt Lake City, Utah, USA.
references
Articles from Annals of Oncology are provided here courtesy of Oxford University Press
Full text links
Read article at publisher's site: https://doi.org/10.1093/annonc/mdv278
Read article for free, from open access legal sources, via Unpaywall:
http://www.annalsofoncology.org/article/S0923753419357837/pdf
Citations & impact
Impact metrics
Citations of article over time
Alternative metrics
Smart citations by scite.ai
Explore citation contexts and check if this article has been
supported or disputed.
https://scite.ai/reports/10.1093/annonc/mdv278
Article citations
In Silico Prediction of BRCA1 and BRCA2 Variants with Conflicting Clinical Interpretation in a Cohort of Breast Cancer Patients.
Genes (Basel), 15(7):943, 18 Jul 2024
Cited by: 0 articles | PMID: 39062721 | PMCID: PMC11276437
Evolving Management of Breast Cancer in the Era of Predictive Biomarkers and Precision Medicine.
J Pers Med, 14(7):719, 03 Jul 2024
Cited by: 1 article | PMID: 39063972 | PMCID: PMC11278458
Review
Free full text in Europe PMC
The Development and Evaluation of Novel Patient Educational Material for a Variant of Uncertain Significance (VUS) Result in Hereditary Cancer Genes.
Curr Oncol, 31(6):3361-3378, 16 Jun 2024
Cited by: 0 articles | PMID: 38920739 | PMCID: PMC11202617
A Systematic Review of the Prevalence of Germline BRCA mutations in North Asia Breast Cancer Patients.
Asian Pac J Cancer Prev, 25(6):1891-1902, 01 Jun 2024
Cited by: 0 articles | PMID: 38918649 | PMCID: PMC11382858
Review
Assessing the Impact of Novel BRCA1 Exon 11 Variants on Pre-mRNA Splicing.
Cells, 13(10):824, 11 May 2024
Cited by: 0 articles | PMID: 38786046 | PMCID: PMC11119505
Go to all (114) article citations
Data
Data behind the article
This data has been text mined from the article, or deposited into data resources.
BioStudies: supplemental material and supporting data
Nucleotide Sequences (2)
- (1 citation) ENA - U14680
- (1 citation) ENA - PMC29713
Similar Articles
To arrive at the top five similar articles we use a word-weighted algorithm to compare words from the Title and Abstract of each citation.
ENIGMA--evidence-based network for the interpretation of germline mutant alleles: an international initiative to evaluate risk and clinical significance associated with sequence variation in BRCA1 and BRCA2 genes.
Hum Mutat, 33(1):2-7, 03 Nov 2011
Cited by: 203 articles | PMID: 21990146 | PMCID: PMC3240687
Reinterpretation of BRCA1 and BRCA2 variants of uncertain significance in patients with hereditary breast/ovarian cancer using the ACMG/AMP 2015 guidelines.
Breast Cancer, 26(4):510-519, 06 Feb 2019
Cited by: 25 articles | PMID: 30725392
Reclassification of BRCA1 and BRCA2 variants of uncertain significance: a multifactorial analysis of multicentre prospective cohort.
J Med Genet, 55(12):794-802, 10 Nov 2018
Cited by: 18 articles | PMID: 30415210
Twenty Years of BRCA1 and BRCA2 Molecular Analysis at MMCI - Current Developments for the Classification of Variants.
Klin Onkol, 32(supplementum2):51-71, 01 Jan 2019
Cited by: 6 articles | PMID: 31409081
Review