Advertisement

A Systematic Comparison of Traditional and Multigene Panel Testing for Hereditary Breast and Ovarian Cancer Genes in More Than 1000 Patients

Open ArchivePublished:July 21, 2015DOI:https://doi.org/10.1016/j.jmoldx.2015.04.009
      Gene panels for hereditary breast and ovarian cancer risk assessment are gaining acceptance, even though the clinical utility of these panels is not yet fully defined. Technical questions remain, however, about the performance and clinical interpretation of gene panels in comparison with traditional tests. We tested 1105 individuals using a 29-gene next-generation sequencing panel and observed 100% analytical concordance with traditional and reference data on >750 comparable variants. These 750 variants included technically challenging classes of sequence and copy number variation that together represent a significant fraction (13.4%) of the pathogenic variants observed. For BRCA1 and BRCA2, we also compared variant interpretations in traditional reports to those produced using only non-proprietary resources and following criteria based on recent (2015) guidelines. We observed 99.8% net report concordance, albeit with a slightly higher variant of uncertain significance rate. In 4.5% of BRCA-negative cases, we uncovered pathogenic variants in other genes, which appear clinically relevant. Previously unseen variants requiring interpretation accumulated rapidly, even after 1000 individuals had been tested. We conclude that next-generation sequencing panel testing can provide results highly comparable to traditional testing and can uncover potentially actionable findings that may be otherwise missed. Challenges remain for the broad adoption of panel tests, some of which will be addressed by the accumulation of large public databases of annotated clinical variants.
      Multigene panel testing has proved useful as a diagnostic tool for disorders where similar phenotypes can be influenced by multiple genes.
      • Rehm H.L.
      Disease-targeted sequencing: a cornerstone in the clinic.
      Recent advances in next-generation DNA sequencing technology (NGS) have enabled these clinical tests and made them increasingly inexpensive to perform.
      • Tucker T.
      • Marra M.
      • Friedman J.M.
      Massively parallel sequencing: the next big thing in genetic medicine.
      • Johansen Taber K.A.
      • Dickinson B.D.
      • Wilson M.
      The promise and challenges of next-generation genome sequencing for clinical care.
      For hereditary cancer syndromes, studies have shown that NGS-based panel tests can uncover potentially actionable findings that may be missed by traditional testing paradigms.
      • Kurian A.W.
      • Hare E.E.
      • Mills M.A.
      • Kingham K.E.
      • McPherson L.
      • Whittemore A.S.
      • McGuire V.
      • Ladabaum U.
      • Kobayashi Y.
      • Lincoln S.E.
      • Cargill M.
      • Ford J.M.
      Clinical evaluation of a multiple-gene sequencing panel for hereditary cancer risk assessment.
      • Tung N.
      • Battelli C.
      • Allen B.
      • Kaldate R.
      • Bhatnagar S.
      • Bowles K.
      • Timms K.
      • Garber J.E.
      • Herold C.
      • Ellisen L.
      • Krejdovsky J.
      • DeLeonardis K.
      • Sedgwick K.
      • Soltis K.
      • Roa B.
      • Wenstrup R.J.
      • Hartman A.R.
      Frequency of mutations in individuals with breast cancer referred for BRCA1 and BRCA2 testing using next-generation sequencing with a 25-gene panel.
      • LaDuca H.
      • Stuenkel A.J.
      • Dolinsky J.S.
      • Keiles S.
      • Tandy S.
      • Pesaran T.
      • Chen E.
      • Gau C.L.
      • Palmaer E.
      • Shoaepour K.
      • Shah D.
      • Speare V.
      • Gandomi S.
      • Chao E.
      Utilization of multigene panels in hereditary cancer predisposition testing: analysis of more than 2,000 patients.
      • Maxwell K.N.
      • Wubbenhorst B.
      • D'Andrea K.
      • Garman B.
      • Long J.M.
      • Powers J.
      • Rathbun K.
      • Stopfer J.E.
      • Zhu J.
      • Bradbury A.R.
      • Simon M.S.
      • DeMichele A.
      • Domchek S.N.
      • Nathanson K.L.
      Prevalence of mutations in a panel of breast cancer susceptibility genes in BRCA1/2-negative patients with early-onset breast cancer.
      • Couch F.J.
      • Hart S.N.
      • Sharma P.
      • Toland A.E.
      • Wang X.
      • Miron P.
      • et al.
      Inherited mutations in 17 breast cancer susceptibility genes among a large triple-negative breast cancer cohort unselected for family history of breast cancer.
      • Minion L.E.
      • Dolinsky J.S.
      • Chase D.M.
      • Dunlop C.L.
      • Chao E.C.
      • Monk B.J.
      Hereditary predisposition to ovarian cancer, looking beyond BRCA1/BRCA2.
      • Walsh T.
      • Casadei S.
      • Lee M.K.
      • Pennil C.C.
      • Nord A.S.
      • Thornton A.M.
      • Roeb W.
      • Agnew K.J.
      • Stray S.M.
      • Wickramanayake A.
      • Norquist B.
      • Pennington K.P.
      • Garcia R.L.
      • King M.-C.
      • Swisher E.M.
      Mutations in 12 genes for inherited ovarian, fallopian tube, and peritoneal carcinoma identified by massively parallel sequencing.
      • Castéra L.
      • Krieger S.
      • Rousselin A.
      • Legros A.
      • Baumann J.-J.
      • Bruet O.
      • Brault B.
      • Fouillet R.
      • Goardon N.
      • Letac O.
      • Baert-Desurmont S.
      • Tinat J.
      • Bera O.
      • Dugast C.
      • Berthet P.
      • Polycarpe F.
      • Layet V.
      • Hardouin A.
      • Frébourg T.
      • Vaur D.
      Next-generation sequencing for the diagnosis of hereditary breast and ovarian cancer using genomic capture targeting multiple candidate genes.
      • Cragun D.
      • Radford C.
      • Dolinsky J.S.
      • Caldwell M.
      • Chao E.
      • Pal T.
      Panel-based testing for inherited colorectal cancer: a descriptive study of clinical testing performed by a US laboratory.
      Validation studies of clinical NGS assays for hereditary cancer genes have correspondingly been published,
      • Kurian A.W.
      • Hare E.E.
      • Mills M.A.
      • Kingham K.E.
      • McPherson L.
      • Whittemore A.S.
      • McGuire V.
      • Ladabaum U.
      • Kobayashi Y.
      • Lincoln S.E.
      • Cargill M.
      • Ford J.M.
      Clinical evaluation of a multiple-gene sequencing panel for hereditary cancer risk assessment.
      • Maxwell K.N.
      • Wubbenhorst B.
      • D'Andrea K.
      • Garman B.
      • Long J.M.
      • Powers J.
      • Rathbun K.
      • Stopfer J.E.
      • Zhu J.
      • Bradbury A.R.
      • Simon M.S.
      • DeMichele A.
      • Domchek S.N.
      • Nathanson K.L.
      Prevalence of mutations in a panel of breast cancer susceptibility genes in BRCA1/2-negative patients with early-onset breast cancer.
      • Castéra L.
      • Krieger S.
      • Rousselin A.
      • Legros A.
      • Baumann J.-J.
      • Bruet O.
      • Brault B.
      • Fouillet R.
      • Goardon N.
      • Letac O.
      • Baert-Desurmont S.
      • Tinat J.
      • Bera O.
      • Dugast C.
      • Berthet P.
      • Polycarpe F.
      • Layet V.
      • Hardouin A.
      • Frébourg T.
      • Vaur D.
      Next-generation sequencing for the diagnosis of hereditary breast and ovarian cancer using genomic capture targeting multiple candidate genes.
      • Chong H.K.
      • Wang T.
      • Lu H.M.
      • Seidler S.
      • Lu H.
      • Keiles S.
      • Chao E.C.
      • Stuenkel A.J.
      • Li X.
      • Elliott A.M.
      The validation and clinical implementation of BRCAplus: a comprehensive high-risk breast cancer diagnostic assay.
      • Bosdet I.E.
      • Docking T.R.
      • Butterfield Y.S.
      • Mungall A.J.
      • Zeng T.
      • Coope R.J.
      • Yorida E.
      • Chow K.
      • Bala M.
      • Young S.S.
      • Hirst M.
      • Birol I.
      • Moore R.A.
      • Jones S.J.
      • Marra M.A.
      • Holt R.
      • Karsan A.
      A clinically validated diagnostic second-generation sequencing assay for detection of hereditary BRCA1 and BRCA2 mutations.
      and certain guidelines exist for their clinical implementation.
      • Stoffel E.M.
      • Mangu P.B.
      • Gruber S.B.
      • Hamilton S.R.
      • Kalady M.F.
      • Lau M.W.Y.
      • Lu K.H.
      • Roach N.
      • Limburg P.J.
      Hereditary colorectal cancer syndromes: American Society of Clinical Oncology Clinical Practice Guideline endorsement of the familial risk–colorectal cancer: European Society for Medical Oncology Clinical Practice Guidelines.
      • Gargis A.S.
      • Kalman L.
      • Berry M.W.
      • Bick D.P.
      • Dimmock D.P.
      • Hambuch T.
      • et al.
      Assuring the quality of next-generation sequencing in clinical laboratory practice.
      • Rehm H.L.
      • Bale S.J.
      • Bayrak-Toydemir P.
      • Berg J.S.
      • Brown K.K.
      • Deignan J.L.
      • Friez M.J.
      • Funke B.H.
      • Hegde M.R.
      • Lyon E.
      Working Group of the American College of Medical Genetics and Genomics Laboratory Quality Assurance Committee
      Hartman clinical laboratory standards for next-generation sequencing.
      • Hall M.J.
      • Forman A.D.
      • Pilarski R.
      • Wiesner G.
      • Giri V.N.
      Gene panel testing for inherited cancer risk.
      Patient management experience using these hereditary cancer panels is growing,
      • Kurian A.W.
      • Hare E.E.
      • Mills M.A.
      • Kingham K.E.
      • McPherson L.
      • Whittemore A.S.
      • McGuire V.
      • Ladabaum U.
      • Kobayashi Y.
      • Lincoln S.E.
      • Cargill M.
      • Ford J.M.
      Clinical evaluation of a multiple-gene sequencing panel for hereditary cancer risk assessment.
      • Mauer C.B.
      • Pirzadeh-Miller S.M.
      • Robinson L.D.
      • Euhus D.M.
      The integration of next-generation sequencing panels in the clinical cancer genetics practice: an institutional experience.
      • Fecteau H.
      • Vogel K.J.
      • Hanson K.
      • Morrill-Cornelius S.
      The evolution of cancer risk assessment in the era of next generation sequencing.
      although the clinical utility of these panels is not yet fully established
      • Robson M.
      Multigene panel testing: planning the next generation of research studies in clinical cancer genetics.

      Grady D, Pollack A: Finding Risks, Not Answers, in Expanding Array of Gene Tests. New York Times 2014 Sep 23, p. A17

      and the appropriate routes for clinical deployment of such tests remain under discussion.
      • Rahman N.
      Mainstreaming genetic testing of cancer predisposition genes.
      Several technical questions also remain about these new tests. NGS has traditionally had analytical limitations
      • Harismendy O.
      • Ng P.C.
      • Strausberg R.L.
      • Wang X.
      • Stockwell T.B.
      • Beeson K.Y.
      • Schork N.J.
      • Murray S.S.
      • Topol E.J.
      • Levy S.
      • Frazer K.A.
      Evaluation of next generation sequencing platforms for population targeted sequencing studies.
      compared with established technologies, such as Sanger sequencing,
      • Sanger F.
      • Nicklen S.
      • Coulson A.R.
      DNA sequencing with chain-terminating inhibitors.
      quantitative PCR,
      • Barrois M.
      • Bièche I.
      • Mazoyer S.
      • Champème M.H.
      • Bressac-de Paillerets B.
      • Lidereau R.
      Real-time PCR-based gene dosage assay for detecting BRCA1 rearrangements in breast–ovarian cancer families.
      multiplex ligation-dependent probe amplification (MLPA),
      • Hogervorst F.B.
      • Nederlof P.M.
      • Gille J.J.
      • McElgunn C.J.
      • Grippeling M.
      • Pruntel R.
      • Regnerus R.
      • van Welsem T.
      • van Spaendonk R.
      • Menko F.H.
      • Kluijt I.
      • Dommering C.
      • Verhoef S.
      • Schouten J.P.
      • van't Veer L.J.
      • Pals G.
      Large genomic deletions and duplications in the BRCA1 gene identified by a novel quantitative method.
      and copy number microarrays.
      • Frolov A.
      • Prowse A.H.
      • Vanderveer L.
      • Bove B.
      • Wu H.
      • Godwin A.K.
      DNA array-based method for detection of large rearrangements in the BRCA1 gene.
      If panel tests are to replace traditional single-gene tests in appropriate situations, further evidence is required to show that NGS can meet the analytic performance standards of these established methods, particularly on those classes of variants that are known to be most challenging for NGS. In addition, questions have long been raised about the potential for inconsistent variant interpretations between laboratories because of limited access to proprietary data and because of differences in interpretation criteria.
      • Plon S.E.
      • Eccles D.M.
      • Easton D.
      • Foulkes W.D.
      • Genuardi M.
      • Greenblatt M.S.
      • Hogervorst F.B.
      • Hoogerbrugge N.
      • Spurdle A.B.
      • Tavtigian S.V.
      IARC Unclassified Genetic Variants Working Group
      Sequence variant classification and reporting: recommendations for improving the interpretation of cancer susceptibility genetic test results.
      • Cook-Deegan R.
      • Conley J.M.
      • Evans J.P.
      • Vorhaus D.
      The next controversy in genetic testing: clinical data as trade secrets?.
      This is an increasingly relevant area for study, because NGS-based tests from multiple laboratories have emerged in recent years
      • Cook-Deegan R.
      • Niehaus A.
      After myriad: genetic testing in the wake of recent Supreme Court decisions about gene patents.
      and new guidelines for the interpretation of sequence variants (ISV) have also emerged.
      • Richards S.
      • Aziz N.
      • Bale S.
      • Bick D.
      • Das S.
      • Gastier-Foster J.
      • Grody W.W.
      • Hegde M.
      • Lyon E.
      • Spector E.
      • Voelkerding K.
      • Rehm H.L.
      Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology.
      To help address these technical questions, particularly as they apply to hereditary breast/ovarian cancer (HBOC), we tested 1062 patients with an NGS-based 29-gene hereditary cancer panel. These individuals were indicated for HBOC risk assessment under National Comprehensive Cancer Network (NCCN) guidelines,
      • Daly M.B.
      • Pilarski R.
      • Axilbund J.E.
      • Buys S.S.
      • Crawford B.
      • Friedman S.
      • Garber J.E.
      • Horton C.
      • Kaklamani V.
      • Klein C.
      • Kohlmann W.
      • Kurian A.
      • Litton J.
      • Madlensky L.
      • Marcom P.K.
      • Merajver S.D.
      • Offit K.
      • Pal T.
      • Pasche B.
      • Reiser G.
      • Shannon K.M.
      • Swisher E.
      • Voian N.C.
      • Weitzel J.N.
      • Whelan A.
      • Wiesner G.L.
      • Dwyer M.A.
      • Kumar R.
      Genetic/familial high-risk assessment: breast and ovarian, version 1.2014.
      and most had previously received clinical testing for BRCA1 and/or BRCA2 from another laboratory. We supplemented these data with additional confirmatory testing and with data from 43 additional reference samples to evaluate NGS performance over more genes and variants. We classified variants following a system based on the recent ISV guidelines, and we used only broadly available, non-proprietary resources to do this. In total, these data allowed us to compare both the analytical and clinical interpretation results from traditional BRCA testing with an NGS-based gene panel.

      Materials and Methods

       Patients, Samples, and Previous Test Data

      Samples were compiled from multiple sources (Table 1 and Supplemental Table S1), and most had previous genetic data available for comparison (Supplemental Table S2).
      Table 1Study Population
      PopulationGroupNo. of patientsDescriptionPrevious testing
      Clinical cases

      (n = 1062)
      Clinical referral735Patients prospectively accrued following NCCN guidelines for HBOCTraditional clinical testing for BRCA1 and/or BRCA2 in most cases, with occasional testing for other genes
      History enriched

      (n = 327)
      209Retrospective cases from a clinical biobank containing particularly high-risk patients
      118Cases referred because of a known pathogenic variant in familyClinical single-site testing
      Reference samples

      (n = 43)
      Positive reference samples36Reference samples selected from public biobanksSamples carry known pathogenic variants in specific genes
      Genome reference samples7Reference samples from public biobanks with high-quality WGS dataVariants in 29 cancer genes extracted from WGS data
      Total1105
      Individuals included in this study, with their selection criteria. Previous test results for BRCA1 and/or BRCA2 were available for 92% (n = 975) of the 1062 clinical cases. All of the reference samples had previous test data. The specific type and scope of testing varied (Supplemental Table S2).
      HBOC, hereditary breast/ovarian cancer; NCCN, National Comprehensive Cancer Network; WGS, whole genome sequencing.
      Seven-hundred thirty-five patients referred for HBOC counseling and/or testing under NCCN guidelines
      • Daly M.B.
      • Pilarski R.
      • Axilbund J.E.
      • Buys S.S.
      • Crawford B.
      • Friedman S.
      • Garber J.E.
      • Horton C.
      • Kaklamani V.
      • Klein C.
      • Kohlmann W.
      • Kurian A.
      • Litton J.
      • Madlensky L.
      • Marcom P.K.
      • Merajver S.D.
      • Offit K.
      • Pal T.
      • Pasche B.
      • Reiser G.
      • Shannon K.M.
      • Swisher E.
      • Voian N.C.
      • Weitzel J.N.
      • Whelan A.
      • Wiesner G.L.
      • Dwyer M.A.
      • Kumar R.
      Genetic/familial high-risk assessment: breast and ovarian, version 1.2014.
      were prospectively recruited at two academic medical centers: the Stanford Clinical Cancer Genetics Program (Stanford, CA; 2002-2012) and the Massachusetts General Hospital Center for Cancer Risk Assessment (Boston, MA; 2013-2014). A further 118 patients referred to either center because of known familial mutations were recruited but considered separately in this analysis. An additional group of 209 patients was recruited at Massachusetts General Hospital Center for Cancer Risk Assessment (2000-2012) on the basis of high-risk personal and family features, but not under uniform criteria, and they are also considered separately. Of the total 1062 patients, 975 (92%) had previously received traditional BRCA1 and/or BRCA2 tests from Myriad Genetics (Salt Lake City, UT), and a small subset (4%) had undergone tests for other genes or multigene panels, as had been clinically indicated. A subset of these patients (n = 175) had been analyzed in our prior work using a research panel,
      • Kurian A.W.
      • Hare E.E.
      • Mills M.A.
      • Kingham K.E.
      • McPherson L.
      • Whittemore A.S.
      • McGuire V.
      • Ladabaum U.
      • Kobayashi Y.
      • Lincoln S.E.
      • Cargill M.
      • Ford J.M.
      Clinical evaluation of a multiple-gene sequencing panel for hereditary cancer risk assessment.
      although insufficient material remained to retest all 198 patients from that previous study here.
      Thirty-six reference samples carrying known pathogenic variants were selected from two public biobanks: the Coriell Institute (Camden, NJ) and the National Institute for Biological Standards and Control (Hertfordshire, UK) (Supplemental Table S3).
      Seven additional reference samples with publicly available whole genome sequence (WGS) data were selected (Supplemental Figure S1). One was NA12878, the pilot reference standard developed by the Genome in a Bottle consortium (National Institute for Standards and Technology, Gaithersburg, MD; http://www.genomeinabottle.org, last accessed February 19, 2015).
      • Zook J.M.
      • Chapman B.
      • Wang J.
      • Mittelman D.
      • Hofmann O.
      • Hide W.
      • Salit M.
      Integrating human sequence data sets provides a resource of benchmark SNP and indel genotype calls.
      All seven came from fully sequenced pedigrees and had data from multiple WGS platforms. Thus, high-accuracy reference calls could be constructed by both integrating the different WGS data sets and using mendelian analysis with phasing.
      • Zook J.M.
      • Chapman B.
      • Wang J.
      • Mittelman D.
      • Hofmann O.
      • Hide W.
      • Salit M.
      Integrating human sequence data sets provides a resource of benchmark SNP and indel genotype calls.
      • Abecasis G.R.
      • Cherny S.S.
      • Cookson W.O.
      • Cardon L.R.
      Merlin: rapid analysis of dense genetic maps using sparse gene flow trees.
      • Roach J.C.
      • Glusman G.
      • Smit A.F.
      • Huff C.D.
      • Hubley R.
      • Shannon P.T.
      • Rowen L.
      • Pant K.P.
      • Goodman N.
      • Bamshad M.
      • Shendure J.
      • Drmanac R.
      • Jorde L.B.
      • Hood L.
      • Galas D.J.
      Analysis of genetic inheritance in a family quartet by whole-genome sequencing.

       Institutional Review Board Approvals

      All patients signed informed consents approved by the Institutional Review Boards of Stanford University (Stanford, CA) or the Dana-Farber Harvard Cancer Center (Boston, MA). Reference samples were used under the terms of their material transfer agreements.

       Gene Panel

      Twenty-nine genes (Table 2 and Supplemental Table S4) were selected on the basis of their documented role in cancer risk
      • Chen S.
      • Parmigiani G.
      Meta-analysis of BRCA1 and BRCA2 penetrance.
      • Mavaddat N.
      • Peock S.
      • Frost D.
      • Ellis S.
      • Platte R.
      • Fineberg E.
      • Evans D.G.
      • Izatt L.
      • Eeles R.A.
      • Adlard J.
      • Davidson R.
      • Eccles D.
      • Cole T.
      • Cook J.
      • Brewer C.
      • Tischkowitz M.
      • Douglas F.
      • Hodgson S.
      • Walker L.
      • Porteous M.E.
      • Morrison P.J.
      • Side L.E.
      • Kennedy M.J.
      • Houghton C.
      • Donaldson A.
      • Rogers M.T.
      • Dorkins H.
      • Miedzybrodzka Z.
      • Gregory H.
      • Eason J.
      • Barwell J.
      • McCann E.
      • Murray A.
      • Antoniou A.C.
      • Easton D.F.
      EMBRACE
      Cancer risks for BRCA1 and BRCA2 mutation carriers: results from prospective analysis of EMBRACE.
      • Pilarski R.
      • Eng C.
      Will the real Cowden syndrome please stand up (again)? expanding mutational and clinical spectra of the PTEN hamartoma tumour syndrome.
      • Tan M.H.
      • Mester J.L.
      • Ngeow J.
      • Rybicki L.A.
      • Orloff M.S.
      • Eng C.
      Lifetime cancer risks in individuals with germline PTEN mutations.
      • Gonzalez K.D.
      • Noltner K.A.
      • Buzin C.H.
      • Gu D.
      • Wen-Fong C.Y.
      • Nguyen V.Q.
      • Han J.H.
      • Lowstuter K.
      • Longmate J.
      • Sommer S.S.
      • Weitzel J.N.
      Beyond Li Fraumeni syndrome: clinical characteristics of families with p53 germline mutations.
      • Hearle N.
      • Schumacher V.
      • Menko F.H.
      • Olschwang S.
      • Boardman L.A.
      • Gille J.J.
      • Keller J.J.
      • Westerman A.M.
      • Scott R.J.
      • Lim W.
      • Trimbath J.D.
      • Giardiello F.M.
      • Gruber S.B.
      • Offerhaus G.J.
      • de Rooij F.W.
      • Wilson J.H.
      • Hansmann A.
      • Möslein G.
      • Royer-Pokora B.
      • Vogel T.
      • Phillips R.K.
      • Spigelman A.D.
      • Houlston R.S.
      Frequency and spectrum of cancers in the Peutz-Jeghers syndrome.
      • Lim W.
      • Olschwang S.
      • Keller J.J.
      • Westerman A.M.
      • Menko F.H.
      • Boardman L.A.
      • Scott R.J.
      • Trimbath J.
      • Giardiello F.M.
      • Gruber S.B.
      • Gille J.J.
      • Offerhaus G.J.
      • de Rooij F.W.
      • Wilson J.H.
      • Spigelman A.D.
      • Phillips R.K.
      • Houlston R.S.
      Relative frequency and morphology of cancers in STK11 mutation carriers.
      • Kaurah P.
      • MacMillan A.
      • Boyd N.
      • Senz J.
      • De Luca A.
      • Chun N.
      • et al.
      Founder and recurrent CDH1 mutations in families with hereditary diffuse gastric cancer.
      • Pharoah P.D.
      • Guilford P.
      • Caldas C.
      International Gastric Cancer Linkage Consortium
      Incidence of gastric cancer and breast cancer in CDH1 (E-cadherin) mutation carriers from hereditary diffuse gastric cancer families.
      • Rahman N.
      • Seal S.
      • Thompson D.
      • Kelly P.
      • Renwick A.
      • Elliott A.
      • Reid S.
      • Spanova K.
      • Barfoot R.
      • Chagtai T.
      • Jayatilake H.
      • McGuffog L.
      • Hanks S.
      • Evans D.G.
      • Eccles D.
      • Easton D.F.
      • Stratton M.R.
      Breast Cancer Susceptibility Collaboration (UK)
      PALB2, which encodes a BRCA2-interacting protein, is a breast cancer susceptibility gene.
      • Antoniou A.C.
      • Casadei S.
      • Heikkinen T.
      • Barrowdale D.
      • Pylkäs K.
      • Roberts J.
      • et al.
      Breast-cancer risk in families with mutations in PALB2.
      • Weischer M.
      • Bojesen S.E.
      • Tybjaerg-Hansen A.
      • Axelsson C.K.
      • Nordestgaard B.G.
      Increased risk of breast cancer associated with CHEK2∗1100delC.
      • Seal S.
      • Thompson D.
      • Renwick A.
      • Elliott A.
      • Kelly P.
      • Barfoot R.
      • Chagtai T.
      • Jayatilake H.
      • Ahmed M.
      • Spanova K.
      • North B.
      • McGuffog L.
      • Evans D.G.
      • Eccles D.
      • Easton D.F.
      • Stratton M.R.
      • Rahman N.
      Breast Cancer Susceptibility Collaboration (UK)
      Truncating mutations in the Fanconi anemia J gene BRIP1 are low-penetrance breast cancer susceptibility alleles.
      • Meindl A.
      • Hellebrand H.
      • Wiek C.
      • Erven V.
      • Wappenschmidt B.
      • Niederacher D.
      • Freund M.
      • Lichtner P.
      • Hartmann L.
      • Schaal H.
      • Ramser J.
      • Honisch E.
      • Kubisch C.
      • Wichmann H.E.
      • Kast K.
      • Deissler H.
      • Engel C.
      • Müller-Myhsok B.
      • Neveling K.
      • Kiechle M.
      • Mathew C.G.
      • Schindler D.
      • Schmutzler R.K.
      • Hanenberg H.
      Germline mutations in breast and ovarian cancer pedigrees establish RAD51C as a human cancer susceptibility gene.
      • Bogdanova N.
      • Feshchenko S.
      • Schürmann P.
      • Waltes R.
      • Wieland B.
      • Hillemanns P.
      • Rogov Y.I.
      • Dammann O.
      • Bremer M.
      • Karstens J.H.
      • Sohn C.
      • Varon R.
      • Dörk T.
      Nijmegen breakage syndrome mutations and risk of breast cancer.
      • Seemanová E.
      • Jarolim P.
      • Seeman P.
      • Varon R.
      • Digweed M.
      • Swift M.
      • Sperling K.
      Cancer risk of heterozygotes with the NBN founder mutation.
      • Ahmed M.
      • Rahman N.
      ATM and breast cancer susceptibility.
      • Win A.K.
      • Lindor N.M.
      • Young J.P.
      • Macrae F.A.
      • Young G.P.
      • Williamson E.
      • Parry S.
      • Goldblatt J.
      • Lipton L.
      • Winship I.
      • Leggett B.
      • Tucker K.M.
      • Giles G.G.
      • Buchanan D.D.
      • Clendenning M.
      • Rosty C.
      • Arnold J.
      • Levine A.J.
      • Haile R.W.
      • Gallinger S.
      • Le Marchand L.
      • Newcomb P.A.
      • Hopper J.L.
      • Jenkins M.A.
      Risks of primary extracolonic cancers following colorectal cancer in lynch syndrome.
      • Jiang L.
      • Zhang C.
      • Li Y.
      • Yu X.
      • Zheng J.
      • Zou P.
      • Li Y.
      • Bin X.
      • Lu J.
      • Zhou Y.
      A non-synonymous polymorphism Thr115Met in the EpCAM gene is associated with an increased risk of breast cancer in Chinese population.
      • Redston M.
      • Nathanson K.L.
      • Yuan Z.Q.
      • Neuhausen S.L.
      • Satagopan J.
      • Wong N.
      • Yang D.
      • Nafa D.
      • Abrahamson J.
      • Ozcelik H.
      • Antin-Ozerkis D.
      • Andrulis I.
      • Daly M.
      • Pinsky L.
      • Schrag D.
      • Gallinger S.
      • Kaback M.
      • King M.C.
      • Woodage T.
      • Brody L.C.
      • Godwin A.
      • Warner E.
      • Weber B.
      • Foulkes W.
      • Offit K.
      The APCI1307K allele and breast cancer risk.
      • Saetrom P.
      • Biesinger J.
      • Li S.M.
      • Smith D.
      • Thomas L.F.
      • Majzoub K.
      • Rivas G.E.
      • Alluin J.
      • Rossi J.J.
      • Krontiris T.G.
      • Weitzel J.
      • Daly M.B.
      • Benson A.B.
      • Kirkwood J.M.
      • O'Dwyer P.J.
      • Sutphen R.
      • Stewart J.A.
      • Johnson D.
      • Larson G.P.
      A risk variant in an miR-125b binding site in BMPR1B is associated with breast cancer pathogenesis.
      • Tram E.
      • Ibrahim-Zada I.
      • Briollais L.
      • Knight J.A.
      • Andrulis I.L.
      • Ozcelik H.
      Identification of germline alterations of the mad homology 2 domain of SMAD3 and SMAD4 from the Ontario site of the breast cancer family registry (CFR).
      • Dean J.L.
      • McClendon A.K.
      • Hickey T.E.
      • Butler L.M.
      • Tilley W.D.
      • Witkiewicz A.K.
      • Knudsen E.S.
      Therapeutic response to CDK4/6 inhibition in breast cancer defined by ex vivo analyses of human tumors.
      • Debniak T.
      • Górski B.
      • Huzarski T.
      • Byrski T.
      • Cybulski C.
      • Mackiewicz A.
      • Gozdecka-Grodecka S.
      • Gronwald J.
      • Kowalska E.
      • Haus O.
      • Grzybowska E.
      • Stawicka M.
      • Swiec M.
      • Urbański K.
      • Niepsuj S.
      • Waśko B.
      • Góźdź S.
      • Wandzel P.
      • Szczylik C.
      • Surdyka D.
      • Rozmiarek A.
      • Zambrano O.
      • Posmyk M.
      • Narod S.A.
      • Lubinski J.
      A common variant of CDKN2A (p16) predisposes to breast cancer.
      • Pogue-Geile K.L.
      • Chen R.
      • Bronner M.P.
      • Crnogorac-Jurcevic T.
      • Moyes K.W.
      • Dowen S.
      • Otey C.A.
      • Crispin D.A.
      • George R.D.
      • Whitcomb D.C.
      • Brentnall T.A.
      Palladin mutation causes familial pancreatic cancer and suggests a new cancer mechanism.
      • Zogopoulos G.
      • Rothenmund H.
      • Eppel A.
      • Ash C.
      • Akbari M.R.
      • Hedley D.
      • Narod S.A.
      • Gallinger S.
      The P239S palladin variant does not account for a significant fraction of hereditary or early onset pancreas cancer.
      • Neklason D.W.
      • Done M.W.
      • Sargent N.R.
      • Schwartz A.G.
      • Anton-Culver H.
      • Griffin C.A.
      • Ahnen D.J.
      • Schildkraut J.M.
      • Tomlinson G.E.
      • Strong L.C.
      • Miller A.R.
      • Stopfer J.E.
      • Burt R.W.
      Activating mutation in MET oncogene in familial colorectal cancer.
      • Lemmens I.
      • Van de Ven W.J.
      • Kas K.
      • Zhang C.X.
      • Giraud S.
      • Wautot V.
      • Buisson N.
      • De Witte K.
      • Salandre J.
      • Lenoir G.
      • Pugeat M.
      • Calender A.
      • Parente F.
      • Quincey D.
      • Gaudray P.
      • De Wit M.J.
      • Lips C.J.
      • Höppener J.W.
      • Khodaei S.
      • Grant A.L.
      • Weber G.
      • Kytölä S.
      • Teh B.T.
      • Farnebo F.
      • Thakker R.V.
      European Consortium on MEN1
      Identification of the multiple endocrine neoplasia type 1 (MEN1) gene.
      • Machens A.
      • Lorenz K.
      • Sekulla C.
      • Höppner W.
      • Frank-Raue K.
      • Raue F.
      • Dralle H.
      Molecular epidemiology of multiple endocrine neoplasia 2: implications for RET screening in the new millenium.
      • Lee S.J.
      • Do I.G.
      • Lee J.
      • Kim K.M.
      • Jang J.
      • Sohn I.
      • Kang W.K.
      Gastric cancer (GC) patients with hedgehog pathway activation: PTCH1 and GLI2 as independent prognostic factors.
      • Kong W.
      • He L.
      • Richards E.J.
      • Challa S.
      • Xu C.X.
      • Permuth-Wey J.
      • Lancaster J.M.
      • Coppola D.
      • Sellers T.A.
      • Djeu J.Y.
      • Cheng J.Q.
      Upregulation of miRNA-155 promotes tumour angiogenesis by targeting VHL and is associated with poor prognosis and triple-negative breast cancer.
      • Rennert G.
      • Lejbkowicz F.
      • Cohen I.
      • Pinchev M.
      • Rennert H.S.
      • Barnett-Griness O.
      MUTYH mutation carriers have increased breast cancer risk.
      and for technical feasibility using the methods described below. These genes were included in a larger NGS assay of 220 genes, most of which are not related to hereditary cancer. We limited our analysis to these 29 genes in this study. Of the 29 genes, 28 are autosomal dominant or have both dominant and recessive cancer phenotypes of varying penetrance. MUTYH was considered separately as a recessive gene, although there is some evidence of elevated cancer risk in monoallelic carriers.
      • Win A.K.
      • Dowty J.G.
      • Cleary S.P.
      • Kim H.
      • Buchanan D.D.
      • Young J.P.
      • Clendenning M.
      • Rosty C.
      • MacInnis R.J.
      • Giles G.G.
      • Boussioutas A.
      • Macrae F.A.
      • Parry S.
      • Goldblatt J.
      • Baron J.A.
      • Burnett T.
      • Le Marchand L.
      • Newcomb P.A.
      • Haile R.W.
      • Hopper J.L.
      • Cotterchio M.
      • Gallinger S.
      • Lindor N.M.
      • Tucker K.M.
      • Winship I.M.
      • Jenkins M.A.
      Risk of colorectal cancer for carriers of mutations in MUTYH, with and without a family history of cancer.
      • Lubbe S.J.
      • Di Bernardo M.C.
      • Chandler I.P.
      • Houlston R.S.
      Clinical implications of the colorectal cancer risk associated with MUTYH mutation.
      The specific version of the assay used in this study had known limitations and did not include exons 12 to 15 of PMS2, because of the known pseudogene (PMS2CL), or regions of CHEK2 other than the c.1100 delC position, because of multiple homologous genomic loci (including the CHEK2P2 pseudogene).
      Table 2List of 29 Genes Tested by Subpanel
      Subpanel nameGenesTotalGene names
      BRCA1/222BRCA1, BRCA2
      Other high-risk breast/ovarian46CDH1, PTEN, STK11, TP53
      Moderate-risk breast/ovarian612ATM, BRIP1, CHEK2, NBN, PALB2, RAD51C
      Lynch syndrome517EPCAM, MLH1, MSH2, MSH6, PMS2
      Other hereditary cancer syndromes1128APC, BMPR1A, SMAD4, CDK4, CDKN2A, PALLD, MET, MEN1, RET, PTCH1, VHL
      MUTYH129MUTYH

       Next Generation Sequencing

      Genomic DNA was extracted from whole blood using a QiaSymphony (Qiagen, Hilden, Germany). Targeted genes were captured using Agilent (Santa Clara, CA) SureSelect probes or Integrated DNA Technologies (Coral, IL) xGen Lockdown probes at positions where SureSelect yield was inadequate. NGS
      • Bentley D.R.
      • Balasubramanian S.
      • Swerdlow H.P.
      • Smith G.P.
      • Milton J.
      • Brown C.G.
      • et al.
      Accurate whole human genome sequencing using reversible terminator chemistry.
      was performed on the Illumina (San Diego, CA) MiSeq or HiSeq 2500 to at least 450× average coverage of 2 × 150 reads, with a minimum of 50× required at every targeted position. Stringent process controls were used to minimize read-depth variability, and up to eight anonymous blood samples were used as control specimens in each run to measure remaining coverage variability.

       Bioinformatics

      Reads were aligned to the reference human genome sequence GRCh37
      • Church D.M.
      • Schneider V.A.
      • Graves T.
      • Auger K.
      • Cunningham F.
      • Bouk N.
      • et al.
      Modernizing reference genome assemblies.
      using Novoalign (Novocraft Technologies, Selangor, Malaysia). Sequence variants were called using a collection of algorithms, including the Genome Analysis Toolkit Unified Genotyper,
      • McKenna A.
      • Hanna M.
      • Banks E.
      • Sivachenko A.
      • Cibulskis K.
      • Kernytsky A.
      • Garimella K.
      • Altshuler D.
      • Gabriel S.
      • Daly M.
      • DePristo M.A.
      The Genome Analysis Toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data.
      Freebayes,

      Garrison E, Marth G: Haplotype-Based Variant Detection from Short-Read Sequencing. arXiv 2012:1207.3907v2 [q-bio.GN] http://arxiv.org/abs/1207.3907

      and Coalgen (unpublished data). Copy number variants (CNVs; alias del/dup events) were called using read-depth analysis by CNVitae.

      Jacobs KB, Paul JS, Nilsen GB, Hart RK, Johnson M, Sorenson J, Cargill M, Lincoln SE: CNVitae–accurate detection of small and large copy number events from DNA capture sequence data. Presented at the Cold Spring Harbor Biology of Genomes Meeting, 2013 May 7-11, Cold Spring Harbor, NY. Abstract 157.

      CNV analysis was not performed on PALLD or MET in this specific assay owing to less predictable read-depth variability. Of 1062 clinical samples, 79 predated the production implementation of CNVitae and, thus, had no NGS CNV assay performed (Supplemental Table S5). Split-read analysis
      • Alkan C.
      • Coe B.P.
      • Eichler E.E.
      Genome structural variation discovery and genotyping.
      was performed in a manner similar to that described elsewhere,
      • Nord A.S.
      • Lee M.
      • King M.-C.
      • Walsh T.
      Accurate and exact CNV identification from targeted high-throughput sequence data.
      although we used this technique at all loci rather than just CNV edges. Split-read analysis allowed us to detect certain structural variants that can evade both read-depth analysis (by having a breakpoint within a small assay target) and indel detection (by being too large or by having a breakpoint outside of the targeted regions).

       Quality Control

      Quality control (QC) parameters included read qualities, mapping qualities, depth of coverage, PCR duplication rates, on/off-target rates, insert size distribution, allelic balance, uniformity of coverage versus controls, and measures of sample contamination (Supplemental Table S6). More than 99% of samples produced data that passed our QC criteria without the need for gap-filling assays, such as Sanger sequencing (Supplemental Table S5). Approximately 10% to 15% of samples required multiple runs to meet these criteria, which we usually performed by repeating both DNA extraction and NGS. Most of these reruns were only required to pass coverage variability metrics related to CNV calling: criteria that apply to the calling of sequence variants were often (>95% samples) met on the first attempt. We found that this rerun rate decreased over time as the technologists gained greater experience with the protocols and as further automation was used in their execution.

       Variant Interpretation

      Updated ISV guidelines were recently established by the American College of Medical Genetics and Genomics and the Association of Molecular Pathology. We used a point-based system based on the last draft of these ISV guidelines.
      • Powers M.P.
      • Anderson M.
      • Garcia J.
      • Nykamp K.
      • Monzon F.
      • Topper S.
      Development of a novel score based system for germline variant interpretation of clinical next-generation sequencing data. Presented at the Association for Molecular Pathology Annual Meeting, 2014 November 12-15, National Harbor, MD.
      That system remains consistent with the final version of these guidelines.
      • Richards S.
      • Aziz N.
      • Bale S.
      • Bick D.
      • Das S.
      • Gastier-Foster J.
      • Grody W.W.
      • Hegde M.
      • Lyon E.
      • Spector E.
      • Voelkerding K.
      • Rehm H.L.
      Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology.
      We classified variants using the five-class system as pathogenic (P), likely pathogenic (LP), variant of uncertain significance (VUS), likely benign (LB), or benign (B). Only non-proprietary data sources were used for this classification, including the published scientific literature, the Universal Mutation Database,
      • Béroud C.
      • Hamroun D.
      • Collod-Béroud G.
      • Boileau C.
      • Soussi T.
      • Claustres M.
      UMD (Universal Mutation Database): 2005 update.
      and the International Society of Gastrointestinal Hereditary Tumours database.
      • Kohonen-Corish M.R.
      • Macrae F.
      • Genuardi M.
      • Aretz S.
      • Bapat B.
      • Bernstein I.T.
      • Burn J.
      • Cotton R.G.
      • den Dunnen J.T.
      • Frebourg T.
      • Greenblatt M.S.
      • Hofstra R.
      • Holinski-Feder E.
      • Lappalainen I.
      • Lindblom A.
      • Maglott D.
      • Møller P.
      • Morreau H.
      • Möslein G.
      • Sijmons R.
      • Spurdle A.B.
      • Tavtigian S.
      • Tops C.M.
      • Weber T.K.
      • de Wind N.
      • Woods M.O.
      Contributors to the InSiGHT-HVP Workshop
      Deciphering the colon cancer genes: report of the InSiGHT-Human Variome Project Workshop, UNESCO, Paris 2010.
      Population databases, such as the 1000 Genomes Project (http://www.1000genomes.org, last accessed November 1, 2014)
      • Abecasis G.R.
      • Auton A.
      • Brooks L.D.
      • DePristo M.A.
      • Durbin R.M.
      • Handsaker R.E.
      • Kang H.M.
      • Marth G.T.
      • McVean G.A.
      1000 Genomes Project Consortium
      An integrated map of genetic variation from 1,092 human genomes.
      and the Exome Variant Server (National Heart, Lung, and Blood Institute GO Exome Sequencing Project, Seattle, WA; http://evs.gs.washington.edu/EVS; last accessed November 1, 2014), were used for allele frequencies.
      The previous clinical tests used proprietary data
      • Cook-Deegan R.
      • Conley J.M.
      • Evans J.P.
      • Vorhaus D.
      The next controversy in genetic testing: clinical data as trade secrets?.
      and somewhat different interpretation criteria.
      • Eggington J.M.
      • Bowles K.R.
      • Moyes K.
      • Manley S.
      • Esterling L.
      • Sizemore S.
      • Rosenthal E.
      • Theisen A.
      • Saam J.
      • Arnell C.
      • Pruss D.
      • Bennett J.
      • Burbidge L.A.
      • Roa B.
      • Wenstrup R.J.
      A comprehensive laboratory-based program for classification of variants of uncertain significance in hereditary cancer genes.
      They also used different nomenclature (eg, deleterious and favor polymorphism), which we mapped onto the ISV five-class nomenclature.

       Orthogonal Confirmation

      Variants were tested using Sanger sequencing
      • Sanger F.
      • Nicklen S.
      • Coulson A.R.
      DNA sequencing with chain-terminating inhibitors.
      or MLPA.
      • Hogervorst F.B.
      • Nederlof P.M.
      • Gille J.J.
      • McElgunn C.J.
      • Grippeling M.
      • Pruntel R.
      • Regnerus R.
      • van Welsem T.
      • van Spaendonk R.
      • Menko F.H.
      • Kluijt I.
      • Dommering C.
      • Verhoef S.
      • Schouten J.P.
      • van't Veer L.J.
      • Pals G.
      Large genomic deletions and duplications in the BRCA1 gene identified by a novel quantitative method.

       Blinding

      NGS sequencing, production bioinformatics, data QC, and clinical interpretation were performed blind to the previous test results. Raw (fastq) data for older specimens in this study were reprocessed after unblinding to produce a consistent data set with uniform bioinformatics and QC (Supplemental Table S5). Interpretations for some variants were updated over the course of this study following an objective process (as is used in clinical practice to amend reports) based solely on the appearance of new evidence or criteria. Laboratory staff with access to the previous test data after unblinding were not involved in this reprocessing or interpretation updates. Newer specimens without reprocessing (n = 313) showed the same degree of concordance as older specimens, suggesting that no bias was introduced.

       Statistical Analysis

      Binomial proportion CIs were calculated using the Jeffreys method. We considered the possible overestimation of performance from discrepant analysis,
      • McAdam A.J.
      Discrepant analysis: how can we test a test?.
      but given that all discordances turned out to be cases where corresponding tests had not been performed (n = 143) or where the reference data were clearly incorrect (n = 3), it was determined to not be a significant concern in this study.

       Data Submission

      De-identified variants and their interpretations from this study have been submitted to ClinVar (National Center for Biotechnology Information, Bethesda, MD; http://www.ncbi.nlm.nih.gov/clinvar, last accessed November 1, 2014).
      • Landrum M.J.
      • Lee J.M.
      • Riley G.R.
      • Jang W.
      • Rubinstein W.S.
      • Church D.M.
      • Maglott D.R.
      ClinVar: public archive of relationships among sequence variation and human phenotype.
      Accession numbers are available in Supplemental Table S7.

      Results

       Spectrum of Variation

      From the 29-gene panel, 58,708 variants were observed across our 1105 individuals. The number of variants in each gene varied considerably and did not strongly correlate with gene size (Supplemental Table S8). The vast majority of these (>99.5%) were single-nucleotide variants, and most (>90%) were common polymorphisms present in the 1000 Genomes Project data set
      • Abecasis G.R.
      • Auton A.
      • Brooks L.D.
      • DePristo M.A.
      • Durbin R.M.
      • Handsaker R.E.
      • Kang H.M.
      • Marth G.T.
      • McVean G.A.
      1000 Genomes Project Consortium
      An integrated map of genetic variation from 1,092 human genomes.
      at a minor allele frequency of 1% or greater.

       Pathogenic Variants

      In 1062 clinical cases, 260 variants classified as P or LP (collectively P/LP) were observed (Figure 1 and Supplemental Table S9). Focusing on the referral cohort (n = 735) as our most clinically representative population, P/LP variants in BRCA1 or BRCA2 (collectively, BRCA) were observed in 9.0% (n = 66) patients, as expected given our criteria for accrual.
      • Hall M.J.
      • Reid J.E.
      • Wenstrup R.J.
      Prevalence of BRCA1 and BRCA2 mutations in women with breast carcinoma in situ and referred for genetic testing.
      Among the BRCA-negative patients, 3.9% (n = 26) carried P/LP variants in other dominant-acting genes. Almost all of these non-BRCA findings were in genes with a known association to breast/ovarian cancer, or were in genes associated with Lynch syndrome. The only exception in the referral cohort was a single patient with a CDKN2A pathogenic variant. The most common non-BRCA findings were in ATM (five cases), PALB2 (five cases), CHEK2 (three cases), and the Lynch syndrome genes (MLH1, MSH2, MSH6, and PMS2; collectively, eight cases). In addition, 2.7% of patients were monoallelic carriers of a P/LP variant in MUTYH, although we saw no biallelic MUTYH carriers (ie, homozygotes or compound heterozygotes). In the history-enriched group, a much larger fraction (40%) of patients were positive for BRCA1 or BRCA2, as expected, and more of the BRCA-negative individuals (6.1%) were positive for other dominant-acting genes.
      Figure thumbnail gr1
      Figure 1Pathogenic variants by subpanel. Number of cases with pathogenic or likely pathogenic variants observed in BRCA1/BRCA2 and in 27 other genes (A) and in these 27 other genes by subpanel (B). The clinical referral cohort represents our best estimate of prevalence in a clinically representative population.
      A small fraction (4%) of the patients had previously received one or more tests for non-BRCA genes (Supplemental Table S2) in addition to (n = 30) or instead of (n = 13) tests for BRCA. These test results were usually negative. Only 3 of 26 non-BRCA P/LP variants in the referral cohort were detected by any previous test: two (in CHEK2 and CDH1, respectively) by a gastrointestinal cancer panel (ColoNext; Ambry Genetics, Aliso Viejo, CA) and one (in PALB2) by a single-gene test. In this study, most non-BRCA findings were detected only through uniform application of the 29-gene panel.
      Some unexpected results were observed among the 118 individuals who had previously received only single-site testing for familial pathogenic variants in BRCA1 or BRCA2. Two, negative for those familial variants, had pathogenic variants detected by the panel in different genes (ATM and MSH2, respectively). These were confirmed and determined unlikely to be the result of sample swaps or clinical record errors; however, we were unable to recontact the two patients to reconfirm these findings in independent specimens. A third individual carried a familial BRCA2 pathogenic variant in addition to a previously undetected pathogenic variant in ATM, whereas a fourth carried a familial BRCA2 pathogenic variant and a previously undetected BRCA1 pathogenic variant.

       Variants of Uncertain Significance

      Among the 1062 clinical cases, 41.0% had at least one VUS among the 29 genes, with 11.4% having two or more VUSs (Figure 2). As expected, the prevalence of VUSs increases considerably with the number of genes tested, and indeed it increases significantly faster than the additional yield of P/LP variants (Figure 1). Most (68%) of these VUSs were rare missense variants not present in the 1000Genomes Project database.
      • Béroud C.
      • Hamroun D.
      • Collod-Béroud G.
      • Boileau C.
      • Soussi T.
      • Claustres M.
      UMD (Universal Mutation Database): 2005 update.
      Figure thumbnail gr2
      Figure 2Prevalence of variants of unknown significance (VUSs). Cumulative fraction of clinical cases with one or more VUSs reported, irrespective of pathogenic variants observed, as the scope of testing increases.

       Technically Challenging Variants for NGS

      A small minority (<0.1%) of the 58,708 variants observed were of types known to be relatively challenging for NGS, defined herein as either large indels (sequence deletions >10 bp or insertions >5 bp); CNVs (alias del/dup events), including gains and losses as small as one exon; block substitutions (delins events); short-range haplotypes (variants within 25 bp); or known homopolymer-associated variants. Given the analytic results described below, we believe this low rate reflects the prevalence of these types of variants more than assay sensitivity limitations.
      In contrast, a strikingly disproportionate fraction (13.4%) of the 260 P/LP variants in the clinical populations were of these challenging types (Figure 3). CNVs were 8.4%, almost half of which (3.8% of the P/LP total) affected only a single exon, consistent with findings reported elsewhere using microarrays.
      • Aradhya S.
      • Lewis R.
      • Bonaga T.
      • Nwokekeh N.
      • Stafford A.
      • Boggs B.
      • Hruska K.
      • Smaoui N.
      • Compton J.G.
      • Richard G.
      • Suchy S.
      Exon-level array CGH in a large clinical cohort demonstrates increased sensitivity of diagnostic testing for mendelian disorders.
      • Retterer K.
      • Scuffins J.
      • Schmidt D.
      • Lewis R.
      • Pineda-Alvarez D.
      • Stafford A.
      • Schmidt L.
      • Warren S.
      • Gibellini F.
      • Kondakova A.
      • Blair A.
      • Bale S.
      • Matyakhina L.
      • Meck J.
      • Aradhya S.
      • Haverfield E.
      Assessing copy number from exome sequencing and exome array CGH based on CNV spectrum in a large clinical cohort.
      Many (15 of 29) of the CNVs were either rare or novel, highlighting the value of unbiased methods to detect these variants, as opposed to limited genotyping assays. The largest sequence-level variant we observed (BRCA2: c.9203del126) was detected only by split-read analysis (Supplemental Figure S2) and not by copy number (CNVitae) or standard indel callers (Genome Analysis Toolkit and Freebayes). The largest sequence insertion we observed was a third copy of a GC-rich 24-bp tandem duplication (CDKN2A: c.9_32dup24), which produced a complex, yet strong, signal by NGS (Supplemental Figure S3).
      Figure thumbnail gr3
      Figure 3Pathogenic variants by type. Pathogenic variants (n = 260) among the clinical cases (n = 1062) by variant type. A significant fraction (13.4%) of the pathogenic variants are of types known to be technically challenging for next-generation sequencing, even though they are a small fraction (<0.1%) of the total set of variants. Single-nucleotide variant (SNV) and copy number variant (CNV) are shown.

       Analytic Validation

      We approached analytic validation by first comparing all results from the previous testing with the 29-gene panel data (Supplemental Figure S4). Any concordant variant was considered a confirmed true positive for the panel, otherwise a false negative (FN). Given limitations in the previous testing, however (Supplemental Table S2), we knew that the reverse comparison would not be as simple: many panel test results would not be present in the previous reports, making it harder to measure false positives. Furthermore, because the previous reports usually omitted variants classified as B or LB, analytic concordance could be confounded with interpretation concordance. We addressed these issues by also orthogonally confirming 143 previously unreported variants, prioritizing those that were P or LP, were technically important (ie, any other indels and CNVs), or helped answer a specific question, in particular whether VUS versus negative report differences were a result of an interpretation difference or an analytic difference. We examined the previous test records and found that all 143 were not previously reported because either a corresponding test had not been previously performed or they were likely filtered from the previous clinical test results by virtue of being classified as benign. Thus, we have no evidence of FNs in the previous test results.
      In total, 750 variants were considered where the 29-gene panel could be directly compared with gold standard data, from either the previous testing or orthogonal confirmation. All 750 were determined to be concordant true positives (Table 3); no false positives or FNs were observed in the 29-gene panel results. More important, these 750 represented a diversity of variant types (Table 4), including 48 of the technically challenging classes listed above (eg, CNVs and large indels). Of these 48, 10 were not previously reported, again because corresponding tests were not previously performed. The 750 also represented a diversity of classifications: 38% were P/LP, 13% VUS, and 49% B/LB, with most of the benign variants contributed by the seven reference genome data sets. Approximately half (372 of 750) of these variants were in genes other than BRCA1 or BRCA2, and all 29 genes except one (CDK4) had at least one variant in this set of 750 (Supplemental Table S8). Most (607 of 750) of these variants were reported by the previous tests, providing the most unbiased view of sensitivity for the NGS panel (607 of 607 = 100%; 95% CI, 100%–99.59%).
      Table 3Analytic Concordance
      29-Gene NGS panelPrevious testing or independent confirmationResult
      Variant presentVariant not present
      Variant detected750 True positives0 False positives100% Sensitivity

      (95% CI for sequence, 100%–99.7%)

      (95% CI for CNV, 100%–91.8%)
      Variant not detected0 False negativesSequence, 15.0 M true negative

      Base pairs

      CNV, 22.2 K true negative exons
      100% Specificity

      (95% CI for Sequence, 100%–99.99998%)

      (95% CI for CNV, 100%–99.989%)
      Concordance of variant calls in all samples (n = 1105) between the 29-gene panel and the independent test results. Variants are counted regardless of pathogenicity if a corresponding test was performed both previously and by the 29-gene panel, or an independent confirmation experiment was performed. The CIs for sensitivity are computed on a per-variant basis; and for specificity, computed on a per–base-pair (sequence variants) or per-exon (CNV) basis.
      CNV, copy number variant; K, thousand; M, million.
      Table 4Variants in Analytic Study
      TypeVariantsSizes or genes
      SequenceSNVs549
      Sequence deletions <10 bp125
      Sequence deletions ≥10 bp9126, 40, 19, 15, 11 bp
      Sequence insertions <5 bp31
      Sequence insertions ≥5 bp424, 5 bp
      Complex variants6Delins, haplotypes, homopolymer associated
      Copy numberSingle-exon deletions9BRCA1, BRCA2, MSH2, PMS2
      Deletions of multiple exons or whole gene10BRCA1, MSH2, RAD51C
      Single-exon duplications4BRCA1, MLH1
      Duplications of multiple exons or whole gene6BRCA1, BRCA2, NBN, SMAD4
      Total750
      Variants in all samples (n = 1105) selected for the analytic validation study. This list includes both variants detected previously and those detected by the 29-gene panel and then subjected to orthogonal confirmation (Supplemental Figure S4). All 750 were confirmed as true positives (Table 3).
      SNV, single-nucleotide variant.
      To estimate the true-negative rate, we counted regions with concordant negative results, disregarding benign variants and counting only those cases for which the 29-gene panel and the previous tests were complete (ie, full-gene sequencing or full copy number assessment was performed). This allowed us to calculate CIs for specificity on a per–base-pair or per-exon basis (Table 3). Calculated instead on a per-test basis, counting each comparable gene by individual combination as one test, the CI for specificity and true-negative count is 100% to 99.86% (n = 1839) for sequence variants and 100% to 99.76% (n = 1026) for CNVs.

       BRCA Interpretations

      Among the 1062 clinical cases, 975 had previous BRCA1 and/or BRCA2 tests that could be compared with the 29-gene panel test results. To determine how often differences in variant interpretation could lead to significantly different test reports, we first categorized patients as BRCA positive (P or LP variant in BRCA1 or BRCA2), uncertain (no P/LP variants but at least one VUS), or BRCA negative (no P/LP variants nor VUSs). Furthermore, because clinicians manage patients with uncertain reports according to personal and family history, and not their genetic test results,
      • Daly M.B.
      • Pilarski R.
      • Axilbund J.E.
      • Buys S.S.
      • Crawford B.
      • Friedman S.
      • Garber J.E.
      • Horton C.
      • Kaklamani V.
      • Klein C.
      • Kohlmann W.
      • Kurian A.
      • Litton J.
      • Madlensky L.
      • Marcom P.K.
      • Merajver S.D.
      • Offit K.
      • Pal T.
      • Pasche B.
      • Reiser G.
      • Shannon K.M.
      • Swisher E.
      • Voian N.C.
      • Weitzel J.N.
      • Whelan A.
      • Wiesner G.L.
      • Dwyer M.A.
      • Kumar R.
      Genetic/familial high-risk assessment: breast and ovarian, version 1.2014.
      we compared positive with not positive (ie, uncertain or negative) reports (Table 5). On this per-patient basis, we see 99.8% concordance (973 of 975; 95% CI, 99.3%–99.9%) between the previous reports and NGS panel test results (Table 5). The two discordant patients had variants reported by the previous laboratory as pathogenic (BRCA1:c.4986+3G>C and BRCA2:c.316+5G>A, respectively), for which there is inadequate publicly available evidence to support pathogenicity (Supplemental Table S10).
      Table 5Interpretation Concordance for BRCA1/BRCA2
      BRCA1/BRCA2 results from 29-gene panelPrevious BRCA1/BRCA2 testing results
      PositiveUncertainNegativeTotal
      Positive18800188

      (19.3%)
      Uncertain2308787

      (80.7%)
      Negative01746
      Total190

      (19.5%)
      785

      (80.5%)
      975
      Concordance of net results for BRCA testing of each patient whenever corresponding tests for BRCA1 or BRCA2 were performed (n = 975 of 1062 patients). Bold numbers are concordant results, and italicized numbers are discordant results. Of 975 cases, 973 (99.8%) are concordant for positive versus nonpositive results. The positive category includes pathogenic and likely pathogenic variants as well as positive single-site and mutation panel tests. The negative category includes patients with only benign or likely benign variants, or with no variants reported. Considering only tests that were capable of reporting variants of uncertain significance, 4.0% (31 of 769) had uncertain reports from the previous testing.
      We further examined specific interpretations for the 1847 comparable BRCA variant observations in these 975 patients. The vast majority (1765 of 1847) had exactly concordant interpretations in the five-class system (Supplemental Table S11), despite the differences in criteria and evidence used. Excluding the two variants mentioned above, the remaining differences had less substantial clinical significance: most differences (65 of 80) were in variants classified as LB by one laboratory and classified as B or entirely omitted from reports by the other. Five were differences between P and LP, and the remaining 10 were differences between VUS and B or LB (Supplemental Table S10). Many of these differences could result from either having different evidence available or differences in the specific details of interpretation criteria. Overall, 95.6% of 1847 had exactly concordant interpretations, and 99.5% were either exactly concordant or one step apart in the five-class system.

      Discussion

      In this study, we compared traditional genetic testing with NGS-based panel testing of 29 hereditary cancer genes in a large population. Most of these individuals had previously received traditional BRCA testing, whereas some had other genetic tests or high-quality reference data available. In this comparison, we were able to consider issues of analytical performance, variant interpretation concordance, and yield of various classes of findings from panel testing.
      In comparing our NGS panel with the previous data, we observed 100% analytical concordance whenever corresponding tests were performed. Although these results are encouraging for the field, they apply only to the specific gene panel, NGS laboratory protocols, and bioinformatics methods we used. Indeed, some NGS studies have reported much lower specificities
      • Chong H.K.
      • Wang T.
      • Lu H.M.
      • Seidler S.
      • Lu H.
      • Keiles S.
      • Chao E.C.
      • Stuenkel A.J.
      • Li X.
      • Elliott A.M.
      The validation and clinical implementation of BRCAplus: a comprehensive high-risk breast cancer diagnostic assay.
      and/or sensitivities.
      • O'Rawe J.
      • Jiang T.
      • Sun G.
      • Wu Y.
      • Wang W.
      • Hu J.
      • Bodily P.
      • Tian L.
      • Hakonarson H.
      • Johnson W.E.
      • Wei Z.
      • Wang K.
      • Lyon G.J.
      Low concordance of multiple variant-calling pipelines: practical implications for exome and genome sequencing.
      Some studies report good NGS sensitivity but include few, if any, of the most challenging variant types.
      • Chong H.K.
      • Wang T.
      • Lu H.M.
      • Seidler S.
      • Lu H.
      • Keiles S.
      • Chao E.C.
      • Stuenkel A.J.
      • Li X.
      • Elliott A.M.
      The validation and clinical implementation of BRCAplus: a comprehensive high-risk breast cancer diagnostic assay.
      • Bosdet I.E.
      • Docking T.R.
      • Butterfield Y.S.
      • Mungall A.J.
      • Zeng T.
      • Coope R.J.
      • Yorida E.
      • Chow K.
      • Bala M.
      • Young S.S.
      • Hirst M.
      • Birol I.
      • Moore R.A.
      • Jones S.J.
      • Marra M.A.
      • Holt R.
      • Karsan A.
      A clinically validated diagnostic second-generation sequencing assay for detection of hereditary BRCA1 and BRCA2 mutations.
      These comparative results highlight the importance of carefully selecting and validating specific methods for each NGS-based test.
      In contrast with the methods used in this study, some laboratories complement NGS with microarrays for CNV analysis.
      • Chong H.K.
      • Wang T.
      • Lu H.M.
      • Seidler S.
      • Lu H.
      • Keiles S.
      • Chao E.C.
      • Stuenkel A.J.
      • Li X.
      • Elliott A.M.
      The validation and clinical implementation of BRCAplus: a comprehensive high-risk breast cancer diagnostic assay.
      • Retterer K.
      • Scuffins J.
      • Schmidt D.
      • Lewis R.
      • Pineda-Alvarez D.
      • Stafford A.
      • Schmidt L.
      • Warren S.
      • Gibellini F.
      • Kondakova A.
      • Blair A.
      • Bale S.
      • Matyakhina L.
      • Meck J.
      • Aradhya S.
      • Haverfield E.
      Assessing copy number from exome sequencing and exome array CGH based on CNV spectrum in a large clinical cohort.
      Our NGS-based CNV detection was highly accurate, even though many of the CNVs present were small (single exons). Clearly, however, comprehensive testing can require a mix of technologies. Two challenging regions (exons 12 to 15 of PMS2 and most of CHEK2) were excluded from this study because of pseudogenes. Methods for addressing such loci by NGS are available
      • Jia H.
      • Guo Y.
      • Zhao W.
      • Wang K.
      Long-range PCR in next-generation sequencing: comparison of six enzymes and evaluation on the MiSeq sequencer.
      but were not implemented in our laboratory in time for this study. In the future, NGS may become a universal assay, but at present, comprehensive testing can require specialized NGS protocols or non-NGS techniques. Laboratories must carefully consider which technologies to implement and must fully understand and disclose limitations of each test they offer.
      We designed the analytic validation portion of this study to include a broad set of genes and variant types. Despite the many individuals we studied, the total number of challenging variants was still small (n = 48), owing to their relative scarcity. Reference samples with accurate WGSs proved useful in assessing our NGS panel across many genes, although unsurprisingly,
      • Zook J.M.
      • Chapman B.
      • Wang J.
      • Mittelman D.
      • Hofmann O.
      • Hide W.
      • Salit M.
      Integrating human sequence data sets provides a resource of benchmark SNP and indel genotype calls.
      • MacArthur D.G.
      • Balasubramanian S.
      • Frankish A.
      • Huang N.
      • Morris J.
      • Walter K.
      • et al.
      A systematic survey of loss-of-function variants in human protein-coding genes.
      only relatively easy variants (single-nucleotide variants and small indels, none of which were pathogenic) were seen in these otherwise unselected individuals. Furthermore, the WGS data have limitations (including coverage gaps, no-calls, and unassayed variant types)
      • Zook J.M.
      • Chapman B.
      • Wang J.
      • Mittelman D.
      • Hofmann O.
      • Hide W.
      • Salit M.
      Integrating human sequence data sets provides a resource of benchmark SNP and indel genotype calls.
      that can render these data less useful for some validation purposes. Reference samples with challenging variants in coding regions of multiple clinically relevant genes would be helpful for future validation studies.
      Most laboratories (including ours) orthogonally confirm NGS findings to reduce the risk of false positives, consistent with established guidelines.
      • Gargis A.S.
      • Kalman L.
      • Berry M.W.
      • Bick D.P.
      • Dimmock D.P.
      • Hambuch T.
      • et al.
      Assuring the quality of next-generation sequencing in clinical laboratory practice.
      The high concordance of our data raises questions about the value of confirmation in some cases, as others have also recently discussed.
      • Strom S.P.
      • Lee H.
      • Das K.
      • Vilain E.
      • Nelson S.F.
      • Grody W.W.
      • Deignan J.L.
      Assessing the necessity of confirmatory testing for exome-sequencing results in a clinical molecular diagnostic laboratory.
      The 2.2014 revision to the NCCN guidelines

      Daly MB, Pilarski R, Axilbund JE, Berry M, Buys SS, Crawford B, et al: National Comprehensive Cancer Network clinical practice guidelines in oncology. Genetic/familial high-risk assessment: breast and ovarian. Version 1.2015. [Internet]. Available at http://www.nccn.org/professionals/physician_gls/pdf/genetics_screening.pdf

      also expresses concern about potential for NGS FNs, which can be harder to study. We observed no FNs within our panel's analytic range in >1105 individuals, although we potentially could have observed analytic FNs if more thorough previous tests had been performed and had uncovered variants missed by our NGS panel. Assay range limitations seem a more likely source of clinical FNs. With the rapid pace of technology and scientific understanding, there is not always detailed consensus on the specific set of genes, reference transcripts, genomic regions, or variant types that should, at a minimum, be assayed for many clinical applications.
      Most of the previous test results in this study came from a single laboratory that used a large proprietary database to aid in variant interpretation.
      • Cook-Deegan R.
      • Conley J.M.
      • Evans J.P.
      • Vorhaus D.
      The next controversy in genetic testing: clinical data as trade secrets?.
      • Cook-Deegan R.
      • Niehaus A.
      After myriad: genetic testing in the wake of recent Supreme Court decisions about gene patents.
      • Eggington J.M.
      • Bowles K.R.
      • Moyes K.
      • Manley S.
      • Esterling L.
      • Sizemore S.
      • Rosenthal E.
      • Theisen A.
      • Saam J.
      • Arnell C.
      • Pruss D.
      • Bennett J.
      • Burbidge L.A.
      • Roa B.
      • Wenstrup R.J.
      A comprehensive laboratory-based program for classification of variants of uncertain significance in hereditary cancer genes.
      Efforts to generate open and peer-reviewed databases, such as ClinVar,
      • Landrum M.J.
      • Lee J.M.
      • Riley G.R.
      • Jang W.
      • Rubinstein W.S.
      • Church D.M.
      • Maglott D.R.
      ClinVar: public archive of relationships among sequence variation and human phenotype.
      may help improve quality and consistency between laboratories. The revised ISV guidelines may also help, although a critical element of medical and scientific judgment will continue to play a vital role in variant interpretation. The 99.8% case-level concordance we observed for BRCA1 and BRCA2 suggests that the ISV guidelines are robust and generally consistent with established practices, and broadly available data are sufficient to correctly interpret most BRCA variants. Nevertheless, the few discordant cases we observed remain vexing, because we cannot scrutinize the underlying evidence of pathogenicity used by the previous laboratory.
      Given such concerns, professional societies, including the American Medical Association, have adopted positions in favor of genetic data sharing (American Medical Association press release, June 18, 2013; http://www.ama-assn.org/ama/pub/news/news/2013/2013-06-18-new-ama-policies-annual-meeting.page). Consortia have formed to facilitate these data-sharing efforts, including ClinGen (http://www.iccg.org/about-the-iccg/clingen), the Global Alliance for Genomics and Health (http://genomicsandhealth.org), and FreeTheData (http://free-the-data.org, all accessed November 1, 2014). We believe it is critical that clinical variants and evidence used in clinical decision making be made available for peer review. All of the variants in this study were thus submitted to ClinVar before this manuscript submission.
      Data sharing will be particularly important to reduce the number of patients who receive uncertain reports from panel tests, particularly because the prevalence of VUSs increases with the number of genes tested. Genetic counseling practices to appropriately communicate these VUSs are becoming established, recognizing that an increased VUS rate is a by-product of increased diagnostic yield.
      • Kurian A.W.
      • Hare E.E.
      • Mills M.A.
      • Kingham K.E.
      • McPherson L.
      • Whittemore A.S.
      • McGuire V.
      • Ladabaum U.
      • Kobayashi Y.
      • Lincoln S.E.
      • Cargill M.
      • Ford J.M.
      Clinical evaluation of a multiple-gene sequencing panel for hereditary cancer risk assessment.
      • Mauer C.B.
      • Pirzadeh-Miller S.M.
      • Robinson L.D.
      • Euhus D.M.
      The integration of next-generation sequencing panels in the clinical cancer genetics practice: an institutional experience.
      • Fecteau H.
      • Vogel K.J.
      • Hanson K.
      • Morrill-Cornelius S.
      The evolution of cancer risk assessment in the era of next generation sequencing.
      Nevertheless, ultimately classifying variants as pathogenic or benign is important for both clinicians and patients. The total number of variants to classify is daunting: in this study, 0.7 new uninterpreted variants were added per patient (Figure 4), and this rate changed only modestly, even after thousands of additional patients were tested in our laboratory for the same genes (data not presented). Larger and more accurate population allele frequency databases will help classify some of these variants,
      • Shearer A.E.
      • Eppsteiner R.W.
      • Booth K.T.
      • Ephraim S.S.
      • Gurrola 2nd, J.
      • Simpson A.
      • Black-Ziegelbein E.A.
      • Joshi S.
      • Ravi H.
      • Giuffre A.C.
      • Happe S.
      • Hildebrand M.S.
      • Azaiez H.
      • Bayazit Y.A.
      • Erdal M.E.
      • Lopez-Escamez J.A.
      • Gazquez I.
      • Tamayo M.L.
      • Gelvez N.Y.
      • Leal G.L.
      • Jalas C.
      • Ekstein J.
      • Yang T.
      • Usami S.
      • Kahrizi K.
      • Bazazzadegan N.
      • Najmabadi H.
      • Scheetz T.E.
      • Braun T.A.
      • Casavant T.L.
      • LeProust E.M.
      • Smith R.J.
      Utilizing ethnic-specific differences in minor allele frequency to recategorize reported pathogenic deafness variants.
      although in a preliminary investigation, we observed that recently available data from 60,706 exomes (Exome Aggregation Consortium, Cambridge, MA; http://exac.broadinstitute.org, last accessed May 5, 2015) would have allowed only a modest fraction of our VUSs to be reclassified under the same allele frequency thresholds. Functional databases will prove even more critical over time,
      • Powers M.P.
      • Anderson M.
      • Garcia J.
      • Nykamp K.
      • Monzon F.
      • Topper S.
      Development of a novel score based system for germline variant interpretation of clinical next-generation sequencing data. Presented at the Association for Molecular Pathology Annual Meeting, 2014 November 12-15, National Harbor, MD.
      • Spurdle A.B.
      • Healey S.
      • Devereau A.
      • Hogervorst F.B.
      • Monteiro A.N.
      • Nathanson K.L.
      • Radice P.
      • Stoppa-Lyonnet D.
      • Tavtigian S.
      • Wappenschmidt B.
      • Couch F.J.
      • Goldgar D.E.
      ENIGMA
      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.
      • Couch F.J.
      • Rasmussen L.J.
      • Hofstra R.
      • Monteiro A.N.
      • Greenblatt M.S.
      • de Wind N.
      IARC Unclassified Genetic Variants Working Group
      Assessment of functional effects of unclassified genetic variants.
      as will registries including Prospective Registry of MultiPlex Testing (https://connect.patientcrossroads.org/?org=prompt, last accessed November 1, 2014) and Inherited Cancer Registry (http://inheritedcancer.net, last accessed November 1, 2014).
      Figure thumbnail gr4
      Figure 4Accumulation of new variants requiring interpretation. Accumulation of new variants in 29 genes requiring clinical interpretation as patients are added (n = 1062). The slope at the end is 0.7 new variants per patient. Variants with high minor-allele frequencies (>3% in the 1000 Genomes Project data) are automatically classified as benign and are not included in this analysis. Data shown are a smoothed average over 10 randomized orders of patients.
      In conclusion, our data show that NGS can achieve high analytic sensitivity and specificity in comparison with traditional genetic testing methods, even for some technically challenging classes of genetic variation that make up a significant fraction of the pathogenic variants in HBOC. Interpretation concordance for BRCA1 and BRCA2 also was high, demonstrating that broadly available resources combined with recent guidelines can produce results similar to those of an established laboratory using a large proprietary database. Thus, gene panels can be a viable replacement for traditional tests in appropriate circumstances. The additional pathogenic variants uncovered by panel testing appeared clinically relevant, albeit with the caveat of identifying many additional VUSs. Of course, ongoing studies to more fully understand the clinical utility of these panels will be important to help evolve guidelines as laboratory technologies, interpretation methods, and public databases continually improve.

      Note Added in Proof

      The patient management implications of the non-BRCA1/2 findings reported herein are detailed in a separate companion article.
      • Desmond A.
      • Kurian A.W.
      • Gabree M.
      • Mills M.A.
      • Anderson M.J.
      • Kobayashi Y.
      • Horick N.
      • Yang S.
      • Shannon K.M.
      • Tung N.
      • Ford J.M.
      • Lincoln S.E.
      • Ellisen L.W.
      Clinical actionability of multi-gene panel testing for hereditary breast and ovarian cancer risk assessment.

      Acknowledgments

      We thank Andrew Schultz and Devika Salunke for sample processing; Curtis Kautzer, Michael Kennemer, Dan Kvitek, Angela Chou, John Major, and the Invitae laboratory staff for NGS process development and data generation; John Garcia, Blanca Herrera, Yuan-Yuan Ho, Keith Nykamp, Nila Patil, and Janita Thusberg for variant interpretation; Raluca Kurtz, Kristen Shannon, and Michele Gabree for genetic counseling expertise; Scott Topper, Swaroop Aradhya, and Krishna Pant for their valuable insights and leadership; Mark Nichol and Katherine Griffin for assistance with this manuscript; and the patients and families who consented to participate in the Massachusetts General Hospital and Stanford research programs.
      S.E.L., A.W.K., J.M.F., and L.W.E. designed and oversaw this study. Y.K. and M.J.A. oversaw variant interpretation and provided scientific insight. S.E.L., S.Y., A.J.D., M.A.M., G.B.N. and K.B.J. performed data management and data analysis. S.E.L., Y.K., S.Y. and F.A.M. drafted the initial manuscript, and all authors reviewed and contributed to the final manuscript.

      Supplemental Data

      • Supplemental Figure S1

        Pedigrees for genome reference samples. Sample names and pedigrees for the genome reference samples. Check marks indicate the seven samples sequenced in this study. Public data from all individuals in the pedigrees were used to refine the reference genotypes for these individuals through mendelian analysis. CEPH, Centre d'Etude du Polymorphisme Humain (Center for the Study of Human Polymorphisms) (Paris, France).

      • Supplemental Figure S2

        BRCA2:c.9203del126. Visualization of soft-clipped bases forming a split-read signal for BRCA2:c.9203del126. The colored portions of reads on the left (1) are soft-clipped bases that align to the reference genome immediately right of the deletion (2). Similarly, the clipped bases on the right (3) align left of the deletion (2). The deleted interval spans most, but not all, of the exon.

      • Supplemental Figure S3

        CDKN2A:c.9_32dup24. Next-generation sequencing detection of third copy of a 24-bp tandem repeat in CDKN2A. The gene is on the reverse strand, and transcription proceeds right to left. Repeat is 5′-GGCTCCATGCTGCTCCCCGCCGCC-3′ and is almost 80% GC. Wild-type copies (1) and (2) span the 5′ Met codon (3). Only a fraction of reads containing the third copy (4) mapped without clipping. Additional reads containing the third copy mapped with soft-clipped bases on both the left (5) and right (6) sides of the two copies in the reference genome.

      • Supplemental Figure S4

        Process used in the analytic validation study. In total, 610 variants were reported by the previous tests, which could be directly compared with the 29-gene panel data. Of these, 607 were matched. All three that did not match came from the whole genome sequence (WGS) data, and all three occurred in two similar low-complexity regions that are known to be problematic in these WGS data sets. After Sanger confirmation and detailed data review, we determined these to be clearly systematic false positives in the WGS data and we removed them from the study. Most of the 58,708 variants detected by the NGS panel were not reported by the previous testing. One hundred forty-three of these variants were selected by the process shown herein. We examined the previous test records and found that all 143 were not previously reported because either a corresponding test had not been previously performed or they were likely filtered from the previous clinical test results by virtue of being classified as benign. These 143 were not immediately considered discordant but were subjected to orthogonal confirmation using traditional techniques (Sanger sequencing and multiplex ligation-dependent probe amplification), and concordance was measured against the confirmation assay results. The asterisk indicates subject to sufficient sample availability for orthogonal testing. CNV, copy number variant; SNV, single-nucleotide variant.

      References

        • Rehm H.L.
        Disease-targeted sequencing: a cornerstone in the clinic.
        Nat Rev Genet. 2013; 14: 295-300
        • Tucker T.
        • Marra M.
        • Friedman J.M.
        Massively parallel sequencing: the next big thing in genetic medicine.
        Am J Hum Genet. 2009; 85: 294-300
        • Johansen Taber K.A.
        • Dickinson B.D.
        • Wilson M.
        The promise and challenges of next-generation genome sequencing for clinical care.
        JAMA Intern Med. 2014; 174: 275-280
        • Kurian A.W.
        • Hare E.E.
        • Mills M.A.
        • Kingham K.E.
        • McPherson L.
        • Whittemore A.S.
        • McGuire V.
        • Ladabaum U.
        • Kobayashi Y.
        • Lincoln S.E.
        • Cargill M.
        • Ford J.M.
        Clinical evaluation of a multiple-gene sequencing panel for hereditary cancer risk assessment.
        J Clin Oncol. 2014; 32: 2001-2009
        • Tung N.
        • Battelli C.
        • Allen B.
        • Kaldate R.
        • Bhatnagar S.
        • Bowles K.
        • Timms K.
        • Garber J.E.
        • Herold C.
        • Ellisen L.
        • Krejdovsky J.
        • DeLeonardis K.
        • Sedgwick K.
        • Soltis K.
        • Roa B.
        • Wenstrup R.J.
        • Hartman A.R.
        Frequency of mutations in individuals with breast cancer referred for BRCA1 and BRCA2 testing using next-generation sequencing with a 25-gene panel.
        Cancer. 2015; 121: 25-33
        • LaDuca H.
        • Stuenkel A.J.
        • Dolinsky J.S.
        • Keiles S.
        • Tandy S.
        • Pesaran T.
        • Chen E.
        • Gau C.L.
        • Palmaer E.
        • Shoaepour K.
        • Shah D.
        • Speare V.
        • Gandomi S.
        • Chao E.
        Utilization of multigene panels in hereditary cancer predisposition testing: analysis of more than 2,000 patients.
        Genet Med. 2014; 16: 830-837
        • Maxwell K.N.
        • Wubbenhorst B.
        • D'Andrea K.
        • Garman B.
        • Long J.M.
        • Powers J.
        • Rathbun K.
        • Stopfer J.E.
        • Zhu J.
        • Bradbury A.R.
        • Simon M.S.
        • DeMichele A.
        • Domchek S.N.
        • Nathanson K.L.
        Prevalence of mutations in a panel of breast cancer susceptibility genes in BRCA1/2-negative patients with early-onset breast cancer.
        Genet Med. 2014; ([Epub ahead of print], doi:10.1038/gim.2014.176)
        • Couch F.J.
        • Hart S.N.
        • Sharma P.
        • Toland A.E.
        • Wang X.
        • Miron P.
        • et al.
        Inherited mutations in 17 breast cancer susceptibility genes among a large triple-negative breast cancer cohort unselected for family history of breast cancer.
        J Clin Oncol. 2015; 33: 304-311
        • Minion L.E.
        • Dolinsky J.S.
        • Chase D.M.
        • Dunlop C.L.
        • Chao E.C.
        • Monk B.J.
        Hereditary predisposition to ovarian cancer, looking beyond BRCA1/BRCA2.
        Gynecol Oncol. 2015; 137: 86-92
        • Walsh T.
        • Casadei S.
        • Lee M.K.
        • Pennil C.C.
        • Nord A.S.
        • Thornton A.M.
        • Roeb W.
        • Agnew K.J.
        • Stray S.M.
        • Wickramanayake A.
        • Norquist B.
        • Pennington K.P.
        • Garcia R.L.
        • King M.-C.
        • Swisher E.M.
        Mutations in 12 genes for inherited ovarian, fallopian tube, and peritoneal carcinoma identified by massively parallel sequencing.
        Proc Natl Acad Sci U S A. 2011; 108: 18032-18037
        • Castéra L.
        • Krieger S.
        • Rousselin A.
        • Legros A.
        • Baumann J.-J.
        • Bruet O.
        • Brault B.
        • Fouillet R.
        • Goardon N.
        • Letac O.
        • Baert-Desurmont S.
        • Tinat J.
        • Bera O.
        • Dugast C.
        • Berthet P.
        • Polycarpe F.
        • Layet V.
        • Hardouin A.
        • Frébourg T.
        • Vaur D.
        Next-generation sequencing for the diagnosis of hereditary breast and ovarian cancer using genomic capture targeting multiple candidate genes.
        Eur J Hum Genet. 2014; 22: 1305-1313
        • Cragun D.
        • Radford C.
        • Dolinsky J.S.
        • Caldwell M.
        • Chao E.
        • Pal T.
        Panel-based testing for inherited colorectal cancer: a descriptive study of clinical testing performed by a US laboratory.
        Clin Genet. 2014; 86: 510-520
        • Chong H.K.
        • Wang T.
        • Lu H.M.
        • Seidler S.
        • Lu H.
        • Keiles S.
        • Chao E.C.
        • Stuenkel A.J.
        • Li X.
        • Elliott A.M.
        The validation and clinical implementation of BRCAplus: a comprehensive high-risk breast cancer diagnostic assay.
        PLoS One. 2014; 9: e97408
        • Bosdet I.E.
        • Docking T.R.
        • Butterfield Y.S.
        • Mungall A.J.
        • Zeng T.
        • Coope R.J.
        • Yorida E.
        • Chow K.
        • Bala M.
        • Young S.S.
        • Hirst M.
        • Birol I.
        • Moore R.A.
        • Jones S.J.
        • Marra M.A.
        • Holt R.
        • Karsan A.
        A clinically validated diagnostic second-generation sequencing assay for detection of hereditary BRCA1 and BRCA2 mutations.
        J Mol Diagn. 2013; 15: 796-809
        • Stoffel E.M.
        • Mangu P.B.
        • Gruber S.B.
        • Hamilton S.R.
        • Kalady M.F.
        • Lau M.W.Y.
        • Lu K.H.
        • Roach N.
        • Limburg P.J.
        Hereditary colorectal cancer syndromes: American Society of Clinical Oncology Clinical Practice Guideline endorsement of the familial risk–colorectal cancer: European Society for Medical Oncology Clinical Practice Guidelines.
        J Clin Oncol. 2015; 33: 209-217
        • Gargis A.S.
        • Kalman L.
        • Berry M.W.
        • Bick D.P.
        • Dimmock D.P.
        • Hambuch T.
        • et al.
        Assuring the quality of next-generation sequencing in clinical laboratory practice.
        Nat Biotechnol. 2012; 30: 1033-1036
        • Rehm H.L.
        • Bale S.J.
        • Bayrak-Toydemir P.
        • Berg J.S.
        • Brown K.K.
        • Deignan J.L.
        • Friez M.J.
        • Funke B.H.
        • Hegde M.R.
        • Lyon E.
        • Working Group of the American College of Medical Genetics and Genomics Laboratory Quality Assurance Committee
        Hartman clinical laboratory standards for next-generation sequencing.
        Genet Med. 2013; 15: 733-747
        • Hall M.J.
        • Forman A.D.
        • Pilarski R.
        • Wiesner G.
        • Giri V.N.
        Gene panel testing for inherited cancer risk.
        J Natl Compr Canc Netw. 2014; 12: 1339-1346
        • Mauer C.B.
        • Pirzadeh-Miller S.M.
        • Robinson L.D.
        • Euhus D.M.
        The integration of next-generation sequencing panels in the clinical cancer genetics practice: an institutional experience.
        Genet Med. 2014; 16: 407-412
        • Fecteau H.
        • Vogel K.J.
        • Hanson K.
        • Morrill-Cornelius S.
        The evolution of cancer risk assessment in the era of next generation sequencing.
        J Genet Couns. 2014; 23: 633-639
        • Robson M.
        Multigene panel testing: planning the next generation of research studies in clinical cancer genetics.
        J Clin Oncol. 2014; 32: 1987-1989
      1. Grady D, Pollack A: Finding Risks, Not Answers, in Expanding Array of Gene Tests. New York Times 2014 Sep 23, p. A17

        • Rahman N.
        Mainstreaming genetic testing of cancer predisposition genes.
        Clin Med. 2014; 14: 436-439
        • Harismendy O.
        • Ng P.C.
        • Strausberg R.L.
        • Wang X.
        • Stockwell T.B.
        • Beeson K.Y.
        • Schork N.J.
        • Murray S.S.
        • Topol E.J.
        • Levy S.
        • Frazer K.A.
        Evaluation of next generation sequencing platforms for population targeted sequencing studies.
        Genome Biol. 2009; 10: R32
        • Sanger F.
        • Nicklen S.
        • Coulson A.R.
        DNA sequencing with chain-terminating inhibitors.
        Proc Natl Acad Sci U S A. 1977; 74: 5463-5467
        • Barrois M.
        • Bièche I.
        • Mazoyer S.
        • Champème M.H.
        • Bressac-de Paillerets B.
        • Lidereau R.
        Real-time PCR-based gene dosage assay for detecting BRCA1 rearrangements in breast–ovarian cancer families.
        Clin Genet. 2004; 65: 131-136
        • Hogervorst F.B.
        • Nederlof P.M.
        • Gille J.J.
        • McElgunn C.J.
        • Grippeling M.
        • Pruntel R.
        • Regnerus R.
        • van Welsem T.
        • van Spaendonk R.
        • Menko F.H.
        • Kluijt I.
        • Dommering C.
        • Verhoef S.
        • Schouten J.P.
        • van't Veer L.J.
        • Pals G.
        Large genomic deletions and duplications in the BRCA1 gene identified by a novel quantitative method.
        Cancer Res. 2003; 63: 1449-1453
        • Frolov A.
        • Prowse A.H.
        • Vanderveer L.
        • Bove B.
        • Wu H.
        • Godwin A.K.
        DNA array-based method for detection of large rearrangements in the BRCA1 gene.
        Cancer. 2002; 35: 232-241
        • Plon S.E.
        • Eccles D.M.
        • Easton D.
        • Foulkes W.D.
        • Genuardi M.
        • Greenblatt M.S.
        • Hogervorst F.B.
        • Hoogerbrugge N.
        • Spurdle A.B.
        • Tavtigian S.V.
        • IARC Unclassified Genetic Variants Working Group
        Sequence variant classification and reporting: recommendations for improving the interpretation of cancer susceptibility genetic test results.
        Hum Mutat. 2008; 29: 1282-1291
        • Cook-Deegan R.
        • Conley J.M.
        • Evans J.P.
        • Vorhaus D.
        The next controversy in genetic testing: clinical data as trade secrets?.
        Eur J Hum Genet. 2013; 21: 585-588
        • Cook-Deegan R.
        • Niehaus A.
        After myriad: genetic testing in the wake of recent Supreme Court decisions about gene patents.
        Curr Genet Med Rep. 2014; 2: 223-241
        • Richards S.
        • Aziz N.
        • Bale S.
        • Bick D.
        • Das S.
        • Gastier-Foster J.
        • Grody W.W.
        • Hegde M.
        • Lyon E.
        • Spector E.
        • Voelkerding K.
        • Rehm H.L.
        Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology.
        Genet Med. 2015; 17: 405-423
        • Daly M.B.
        • Pilarski R.
        • Axilbund J.E.
        • Buys S.S.
        • Crawford B.
        • Friedman S.
        • Garber J.E.
        • Horton C.
        • Kaklamani V.
        • Klein C.
        • Kohlmann W.
        • Kurian A.
        • Litton J.
        • Madlensky L.
        • Marcom P.K.
        • Merajver S.D.
        • Offit K.
        • Pal T.
        • Pasche B.
        • Reiser G.
        • Shannon K.M.
        • Swisher E.
        • Voian N.C.
        • Weitzel J.N.
        • Whelan A.
        • Wiesner G.L.
        • Dwyer M.A.
        • Kumar R.
        Genetic/familial high-risk assessment: breast and ovarian, version 1.2014.
        J Natl Compr Canc Netw. 2014; 12: 1326-1338
        • Zook J.M.
        • Chapman B.
        • Wang J.
        • Mittelman D.
        • Hofmann O.
        • Hide W.
        • Salit M.
        Integrating human sequence data sets provides a resource of benchmark SNP and indel genotype calls.
        Nat Biotechnol. 2014; 32: 246-251
        • Abecasis G.R.
        • Cherny S.S.
        • Cookson W.O.
        • Cardon L.R.
        Merlin: rapid analysis of dense genetic maps using sparse gene flow trees.
        Nat Genet. 2002; 30: 97-101
        • Roach J.C.
        • Glusman G.
        • Smit A.F.
        • Huff C.D.
        • Hubley R.
        • Shannon P.T.
        • Rowen L.
        • Pant K.P.
        • Goodman N.
        • Bamshad M.
        • Shendure J.
        • Drmanac R.
        • Jorde L.B.
        • Hood L.
        • Galas D.J.
        Analysis of genetic inheritance in a family quartet by whole-genome sequencing.
        Science. 2010; 328: 636-639
        • Chen S.
        • Parmigiani G.
        Meta-analysis of BRCA1 and BRCA2 penetrance.
        J Clin Oncol. 2007; 25: 1329-1333
        • Mavaddat N.
        • Peock S.
        • Frost D.
        • Ellis S.
        • Platte R.
        • Fineberg E.
        • Evans D.G.
        • Izatt L.
        • Eeles R.A.
        • Adlard J.
        • Davidson R.
        • Eccles D.
        • Cole T.
        • Cook J.
        • Brewer C.
        • Tischkowitz M.
        • Douglas F.
        • Hodgson S.
        • Walker L.
        • Porteous M.E.
        • Morrison P.J.
        • Side L.E.
        • Kennedy M.J.
        • Houghton C.
        • Donaldson A.
        • Rogers M.T.
        • Dorkins H.
        • Miedzybrodzka Z.
        • Gregory H.
        • Eason J.
        • Barwell J.
        • McCann E.
        • Murray A.
        • Antoniou A.C.
        • Easton D.F.
        • EMBRACE
        Cancer risks for BRCA1 and BRCA2 mutation carriers: results from prospective analysis of EMBRACE.
        J Natl Cancer Inst. 2013; 105: 812-822
        • Pilarski R.
        • Eng C.
        Will the real Cowden syndrome please stand up (again)? expanding mutational and clinical spectra of the PTEN hamartoma tumour syndrome.
        J Med Genet. 2004; 41: 323-326
        • Tan M.H.
        • Mester J.L.
        • Ngeow J.
        • Rybicki L.A.
        • Orloff M.S.
        • Eng C.
        Lifetime cancer risks in individuals with germline PTEN mutations.
        Clin Cancer Res. 2012; 18: 400-407
        • Gonzalez K.D.
        • Noltner K.A.
        • Buzin C.H.
        • Gu D.
        • Wen-Fong C.Y.
        • Nguyen V.Q.
        • Han J.H.
        • Lowstuter K.
        • Longmate J.
        • Sommer S.S.
        • Weitzel J.N.
        Beyond Li Fraumeni syndrome: clinical characteristics of families with p53 germline mutations.
        J Clin Oncol. 2009; 27: 1250-1256
        • Hearle N.
        • Schumacher V.
        • Menko F.H.
        • Olschwang S.
        • Boardman L.A.
        • Gille J.J.
        • Keller J.J.
        • Westerman A.M.
        • Scott R.J.
        • Lim W.
        • Trimbath J.D.
        • Giardiello F.M.
        • Gruber S.B.
        • Offerhaus G.J.
        • de Rooij F.W.
        • Wilson J.H.
        • Hansmann A.
        • Möslein G.
        • Royer-Pokora B.
        • Vogel T.
        • Phillips R.K.
        • Spigelman A.D.
        • Houlston R.S.
        Frequency and spectrum of cancers in the Peutz-Jeghers syndrome.
        Clin Cancer Res. 2006; 12: 3209-3215
        • Lim W.
        • Olschwang S.
        • Keller J.J.
        • Westerman A.M.
        • Menko F.H.
        • Boardman L.A.
        • Scott R.J.
        • Trimbath J.
        • Giardiello F.M.
        • Gruber S.B.
        • Gille J.J.
        • Offerhaus G.J.
        • de Rooij F.W.
        • Wilson J.H.
        • Spigelman A.D.
        • Phillips R.K.
        • Houlston R.S.
        Relative frequency and morphology of cancers in STK11 mutation carriers.
        Gastroenterology. 2004; 126: 1788-1794
        • Kaurah P.
        • MacMillan A.
        • Boyd N.
        • Senz J.
        • De Luca A.
        • Chun N.
        • et al.
        Founder and recurrent CDH1 mutations in families with hereditary diffuse gastric cancer.
        JAMA. 2007; 297: 2360-2372
        • Pharoah P.D.
        • Guilford P.
        • Caldas C.
        • International Gastric Cancer Linkage Consortium
        Incidence of gastric cancer and breast cancer in CDH1 (E-cadherin) mutation carriers from hereditary diffuse gastric cancer families.
        Gastroenterology. 2001; 121: 1348-1353
        • Rahman N.
        • Seal S.
        • Thompson D.
        • Kelly P.
        • Renwick A.
        • Elliott A.
        • Reid S.
        • Spanova K.
        • Barfoot R.
        • Chagtai T.
        • Jayatilake H.
        • McGuffog L.
        • Hanks S.
        • Evans D.G.
        • Eccles D.
        • Easton D.F.
        • Stratton M.R.
        • Breast Cancer Susceptibility Collaboration (UK)
        PALB2, which encodes a BRCA2-interacting protein, is a breast cancer susceptibility gene.
        Nat Genet. 2007; 39: 165-167
        • Antoniou A.C.
        • Casadei S.
        • Heikkinen T.
        • Barrowdale D.
        • Pylkäs K.
        • Roberts J.
        • et al.
        Breast-cancer risk in families with mutations in PALB2.
        N Engl J Med. 2014; 371: 497-506
        • Weischer M.
        • Bojesen S.E.
        • Tybjaerg-Hansen A.
        • Axelsson C.K.
        • Nordestgaard B.G.
        Increased risk of breast cancer associated with CHEK2∗1100delC.
        J Clin Oncol. 2007; 25: 57-63
        • Seal S.
        • Thompson D.
        • Renwick A.
        • Elliott A.
        • Kelly P.
        • Barfoot R.
        • Chagtai T.
        • Jayatilake H.
        • Ahmed M.
        • Spanova K.
        • North B.
        • McGuffog L.
        • Evans D.G.
        • Eccles D.
        • Easton D.F.
        • Stratton M.R.
        • Rahman N.
        • Breast Cancer Susceptibility Collaboration (UK)
        Truncating mutations in the Fanconi anemia J gene BRIP1 are low-penetrance breast cancer susceptibility alleles.
        Nat Genet. 2006; 38: 1239-1241
        • Meindl A.
        • Hellebrand H.
        • Wiek C.
        • Erven V.
        • Wappenschmidt B.
        • Niederacher D.
        • Freund M.
        • Lichtner P.
        • Hartmann L.
        • Schaal H.
        • Ramser J.
        • Honisch E.
        • Kubisch C.
        • Wichmann H.E.
        • Kast K.
        • Deissler H.
        • Engel C.
        • Müller-Myhsok B.
        • Neveling K.
        • Kiechle M.
        • Mathew C.G.
        • Schindler D.
        • Schmutzler R.K.
        • Hanenberg H.
        Germline mutations in breast and ovarian cancer pedigrees establish RAD51C as a human cancer susceptibility gene.
        Nat Genet. 2010; 42: 410-414
        • Bogdanova N.
        • Feshchenko S.
        • Schürmann P.
        • Waltes R.
        • Wieland B.
        • Hillemanns P.
        • Rogov Y.I.
        • Dammann O.
        • Bremer M.
        • Karstens J.H.
        • Sohn C.
        • Varon R.
        • Dörk T.
        Nijmegen breakage syndrome mutations and risk of breast cancer.
        Int J Cancer. 2008; 122: 802-806
        • Seemanová E.
        • Jarolim P.
        • Seeman P.
        • Varon R.
        • Digweed M.
        • Swift M.
        • Sperling K.
        Cancer risk of heterozygotes with the NBN founder mutation.
        J Natl Cancer Inst. 2007; 99: 1875-1880
        • Ahmed M.
        • Rahman N.
        ATM and breast cancer susceptibility.
        Oncogene. 2006; 25: 5906-5911
        • Win A.K.
        • Lindor N.M.
        • Young J.P.
        • Macrae F.A.
        • Young G.P.
        • Williamson E.
        • Parry S.
        • Goldblatt J.
        • Lipton L.
        • Winship I.
        • Leggett B.
        • Tucker K.M.
        • Giles G.G.
        • Buchanan D.D.
        • Clendenning M.
        • Rosty C.
        • Arnold J.
        • Levine A.J.
        • Haile R.W.
        • Gallinger S.
        • Le Marchand L.
        • Newcomb P.A.
        • Hopper J.L.
        • Jenkins M.A.
        Risks of primary extracolonic cancers following colorectal cancer in lynch syndrome.
        J Natl Cancer Inst. 2012; 104: 1363-1372
        • Jiang L.
        • Zhang C.
        • Li Y.
        • Yu X.
        • Zheng J.
        • Zou P.
        • Li Y.
        • Bin X.
        • Lu J.
        • Zhou Y.
        A non-synonymous polymorphism Thr115Met in the EpCAM gene is associated with an increased risk of breast cancer in Chinese population.
        Breast Cancer Res Treat. 2011; 126: 487-495
        • Redston M.
        • Nathanson K.L.
        • Yuan Z.Q.
        • Neuhausen S.L.
        • Satagopan J.
        • Wong N.
        • Yang D.
        • Nafa D.
        • Abrahamson J.
        • Ozcelik H.
        • Antin-Ozerkis D.
        • Andrulis I.
        • Daly M.
        • Pinsky L.
        • Schrag D.
        • Gallinger S.
        • Kaback M.
        • King M.C.
        • Woodage T.
        • Brody L.C.
        • Godwin A.
        • Warner E.
        • Weber B.
        • Foulkes W.
        • Offit K.
        The APCI1307K allele and breast cancer risk.
        Nat Genet. 1998; 20: 13-14
        • Saetrom P.
        • Biesinger J.
        • Li S.M.
        • Smith D.
        • Thomas L.F.
        • Majzoub K.
        • Rivas G.E.
        • Alluin J.
        • Rossi J.J.
        • Krontiris T.G.
        • Weitzel J.
        • Daly M.B.
        • Benson A.B.
        • Kirkwood J.M.
        • O'Dwyer P.J.
        • Sutphen R.
        • Stewart J.A.
        • Johnson D.
        • Larson G.P.
        A risk variant in an miR-125b binding site in BMPR1B is associated with breast cancer pathogenesis.
        Cancer Res. 2009; 69: 7459-7465
        • Tram E.
        • Ibrahim-Zada I.
        • Briollais L.
        • Knight J.A.
        • Andrulis I.L.
        • Ozcelik H.
        Identification of germline alterations of the mad homology 2 domain of SMAD3 and SMAD4 from the Ontario site of the breast cancer family registry (CFR).
        Breast Cancer Res. 2011; 13: R77
        • Dean J.L.
        • McClendon A.K.
        • Hickey T.E.
        • Butler L.M.
        • Tilley W.D.
        • Witkiewicz A.K.
        • Knudsen E.S.
        Therapeutic response to CDK4/6 inhibition in breast cancer defined by ex vivo analyses of human tumors.
        Cell Cycle. 2012; 11: 2756-2761
        • Debniak T.
        • Górski B.
        • Huzarski T.
        • Byrski T.
        • Cybulski C.
        • Mackiewicz A.
        • Gozdecka-Grodecka S.
        • Gronwald J.
        • Kowalska E.
        • Haus O.
        • Grzybowska E.
        • Stawicka M.
        • Swiec M.
        • Urbański K.
        • Niepsuj S.
        • Waśko B.
        • Góźdź S.
        • Wandzel P.
        • Szczylik C.
        • Surdyka D.
        • Rozmiarek A.
        • Zambrano O.
        • Posmyk M.
        • Narod S.A.
        • Lubinski J.
        A common variant of CDKN2A (p16) predisposes to breast cancer.
        J Med Genet. 2005; 42: 763-765
        • Pogue-Geile K.L.
        • Chen R.
        • Bronner M.P.
        • Crnogorac-Jurcevic T.
        • Moyes K.W.
        • Dowen S.
        • Otey C.A.
        • Crispin D.A.
        • George R.D.
        • Whitcomb D.C.
        • Brentnall T.A.
        Palladin mutation causes familial pancreatic cancer and suggests a new cancer mechanism.
        PLoS Med. 2006; 3: e516
        • Zogopoulos G.
        • Rothenmund H.
        • Eppel A.
        • Ash C.
        • Akbari M.R.
        • Hedley D.
        • Narod S.A.
        • Gallinger S.
        The P239S palladin variant does not account for a significant fraction of hereditary or early onset pancreas cancer.
        Hum Genet. 2007; 121: 635-637
        • Neklason D.W.
        • Done M.W.
        • Sargent N.R.
        • Schwartz A.G.
        • Anton-Culver H.
        • Griffin C.A.
        • Ahnen D.J.
        • Schildkraut J.M.
        • Tomlinson G.E.
        • Strong L.C.
        • Miller A.R.
        • Stopfer J.E.
        • Burt R.W.
        Activating mutation in MET oncogene in familial colorectal cancer.
        BMC Cancer. 2011; 11: 424
        • Lemmens I.
        • Van de Ven W.J.
        • Kas K.
        • Zhang C.X.
        • Giraud S.
        • Wautot V.
        • Buisson N.
        • De Witte K.
        • Salandre J.
        • Lenoir G.
        • Pugeat M.
        • Calender A.
        • Parente F.
        • Quincey D.
        • Gaudray P.
        • De Wit M.J.
        • Lips C.J.
        • Höppener J.W.
        • Khodaei S.
        • Grant A.L.
        • Weber G.
        • Kytölä S.
        • Teh B.T.
        • Farnebo F.
        • Thakker R.V.
        • European Consortium on MEN1
        Identification of the multiple endocrine neoplasia type 1 (MEN1) gene.
        Hum Mol Genet. 1997; 6: 1177-1183
        • Machens A.
        • Lorenz K.
        • Sekulla C.
        • Höppner W.
        • Frank-Raue K.
        • Raue F.
        • Dralle H.
        Molecular epidemiology of multiple endocrine neoplasia 2: implications for RET screening in the new millenium.
        Eur J Endocrinol. 2013; 168: 307-314
        • Lee S.J.
        • Do I.G.
        • Lee J.
        • Kim K.M.
        • Jang J.
        • Sohn I.
        • Kang W.K.
        Gastric cancer (GC) patients with hedgehog pathway activation: PTCH1 and GLI2 as independent prognostic factors.
        Target Oncol. 2013; 8: 271-280
        • Kong W.
        • He L.
        • Richards E.J.
        • Challa S.
        • Xu C.X.
        • Permuth-Wey J.
        • Lancaster J.M.
        • Coppola D.
        • Sellers T.A.
        • Djeu J.Y.
        • Cheng J.Q.
        Upregulation of miRNA-155 promotes tumour angiogenesis by targeting VHL and is associated with poor prognosis and triple-negative breast cancer.
        Oncogene. 2014; 33: 679-689
        • Rennert G.
        • Lejbkowicz F.
        • Cohen I.
        • Pinchev M.
        • Rennert H.S.
        • Barnett-Griness O.
        MUTYH mutation carriers have increased breast cancer risk.
        Cancer. 2012; 118: 1989-1993
        • Win A.K.
        • Dowty J.G.
        • Cleary S.P.
        • Kim H.
        • Buchanan D.D.
        • Young J.P.
        • Clendenning M.
        • Rosty C.
        • MacInnis R.J.
        • Giles G.G.
        • Boussioutas A.
        • Macrae F.A.
        • Parry S.
        • Goldblatt J.
        • Baron J.A.
        • Burnett T.
        • Le Marchand L.
        • Newcomb P.A.
        • Haile R.W.
        • Hopper J.L.
        • Cotterchio M.
        • Gallinger S.
        • Lindor N.M.
        • Tucker K.M.
        • Winship I.M.
        • Jenkins M.A.
        Risk of colorectal cancer for carriers of mutations in MUTYH, with and without a family history of cancer.
        Gastroenterology. 2014; 146: 1208-1211
        • Lubbe S.J.
        • Di Bernardo M.C.
        • Chandler I.P.
        • Houlston R.S.
        Clinical implications of the colorectal cancer risk associated with MUTYH mutation.
        J Clin Oncol. 2009; 27: 3975-3980
        • Bentley D.R.
        • Balasubramanian S.
        • Swerdlow H.P.
        • Smith G.P.
        • Milton J.
        • Brown C.G.
        • et al.
        Accurate whole human genome sequencing using reversible terminator chemistry.
        Nature. 2008; 456: 53-59
        • Church D.M.
        • Schneider V.A.
        • Graves T.
        • Auger K.
        • Cunningham F.
        • Bouk N.
        • et al.
        Modernizing reference genome assemblies.
        PLoS Biol. 2011; 9: e1001091
        • McKenna A.
        • Hanna M.
        • Banks E.
        • Sivachenko A.
        • Cibulskis K.
        • Kernytsky A.
        • Garimella K.
        • Altshuler D.
        • Gabriel S.
        • Daly M.
        • DePristo M.A.
        The Genome Analysis Toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data.
        Genome Res. 2010; 20: 1297-1303
      2. Garrison E, Marth G: Haplotype-Based Variant Detection from Short-Read Sequencing. arXiv 2012:1207.3907v2 [q-bio.GN] http://arxiv.org/abs/1207.3907

      3. Jacobs KB, Paul JS, Nilsen GB, Hart RK, Johnson M, Sorenson J, Cargill M, Lincoln SE: CNVitae–accurate detection of small and large copy number events from DNA capture sequence data. Presented at the Cold Spring Harbor Biology of Genomes Meeting, 2013 May 7-11, Cold Spring Harbor, NY. Abstract 157.

        • Alkan C.
        • Coe B.P.
        • Eichler E.E.
        Genome structural variation discovery and genotyping.
        Nat Rev Genet. 2011; 12: 363-376
        • Nord A.S.
        • Lee M.
        • King M.-C.
        • Walsh T.
        Accurate and exact CNV identification from targeted high-throughput sequence data.
        BMC Genomics. 2011; 12: 184
        • Powers M.P.
        • Anderson M.
        • Garcia J.
        • Nykamp K.
        • Monzon F.
        • Topper S.
        Development of a novel score based system for germline variant interpretation of clinical next-generation sequencing data. Presented at the Association for Molecular Pathology Annual Meeting, 2014 November 12-15, National Harbor, MD.
        J Mol Diagn. 2014; 16 (Abstract G30): 699
        • Béroud C.
        • Hamroun D.
        • Collod-Béroud G.
        • Boileau C.
        • Soussi T.
        • Claustres M.
        UMD (Universal Mutation Database): 2005 update.
        Hum Mutat. 2005; 26: 184-191
        • Kohonen-Corish M.R.
        • Macrae F.
        • Genuardi M.
        • Aretz S.
        • Bapat B.
        • Bernstein I.T.
        • Burn J.
        • Cotton R.G.
        • den Dunnen J.T.
        • Frebourg T.
        • Greenblatt M.S.
        • Hofstra R.
        • Holinski-Feder E.
        • Lappalainen I.
        • Lindblom A.
        • Maglott D.
        • Møller P.
        • Morreau H.
        • Möslein G.
        • Sijmons R.
        • Spurdle A.B.
        • Tavtigian S.
        • Tops C.M.
        • Weber T.K.
        • de Wind N.
        • Woods M.O.
        • Contributors to the InSiGHT-HVP Workshop
        Deciphering the colon cancer genes: report of the InSiGHT-Human Variome Project Workshop, UNESCO, Paris 2010.
        Hum Mutat. 2011; 32: 491-494
        • Abecasis G.R.
        • Auton A.
        • Brooks L.D.
        • DePristo M.A.
        • Durbin R.M.
        • Handsaker R.E.
        • Kang H.M.
        • Marth G.T.
        • McVean G.A.
        • 1000 Genomes Project Consortium
        An integrated map of genetic variation from 1,092 human genomes.
        Nature. 2012; 491: 56-65
        • Eggington J.M.
        • Bowles K.R.
        • Moyes K.
        • Manley S.
        • Esterling L.
        • Sizemore S.
        • Rosenthal E.
        • Theisen A.
        • Saam J.
        • Arnell C.
        • Pruss D.
        • Bennett J.
        • Burbidge L.A.
        • Roa B.
        • Wenstrup R.J.
        A comprehensive laboratory-based program for classification of variants of uncertain significance in hereditary cancer genes.
        Clin Genet. 2014; 86: 229-237
        • McAdam A.J.
        Discrepant analysis: how can we test a test?.
        J Clin Microbiol. 2000; 38: 2027-2029
        • Landrum M.J.
        • Lee J.M.
        • Riley G.R.
        • Jang W.
        • Rubinstein W.S.
        • Church D.M.
        • Maglott D.R.
        ClinVar: public archive of relationships among sequence variation and human phenotype.
        Nucleic Acids Res. 2014; 42: 387-406
        • Hall M.J.
        • Reid J.E.
        • Wenstrup R.J.
        Prevalence of BRCA1 and BRCA2 mutations in women with breast carcinoma in situ and referred for genetic testing.
        Cancer Prev Res (Phila). 2010; 3: 1579-1585
        • Aradhya S.
        • Lewis R.
        • Bonaga T.
        • Nwokekeh N.
        • Stafford A.
        • Boggs B.
        • Hruska K.
        • Smaoui N.
        • Compton J.G.
        • Richard G.
        • Suchy S.
        Exon-level array CGH in a large clinical cohort demonstrates increased sensitivity of diagnostic testing for mendelian disorders.
        Genet Med. 2012; 14: 594-603
        • Retterer K.
        • Scuffins J.
        • Schmidt D.
        • Lewis R.
        • Pineda-Alvarez D.
        • Stafford A.
        • Schmidt L.
        • Warren S.
        • Gibellini F.
        • Kondakova A.
        • Blair A.
        • Bale S.
        • Matyakhina L.
        • Meck J.
        • Aradhya S.
        • Haverfield E.
        Assessing copy number from exome sequencing and exome array CGH based on CNV spectrum in a large clinical cohort.
        Genet Med. 2015; 17: 623-629
        • O'Rawe J.
        • Jiang T.
        • Sun G.
        • Wu Y.
        • Wang W.
        • Hu J.
        • Bodily P.
        • Tian L.
        • Hakonarson H.
        • Johnson W.E.
        • Wei Z.
        • Wang K.
        • Lyon G.J.
        Low concordance of multiple variant-calling pipelines: practical implications for exome and genome sequencing.
        Genome Med. 2013; 5: 387-406
        • Jia H.
        • Guo Y.
        • Zhao W.
        • Wang K.
        Long-range PCR in next-generation sequencing: comparison of six enzymes and evaluation on the MiSeq sequencer.
        Sci Rep. 2014; 4: 387-406
        • MacArthur D.G.
        • Balasubramanian S.
        • Frankish A.
        • Huang N.
        • Morris J.
        • Walter K.
        • et al.
        A systematic survey of loss-of-function variants in human protein-coding genes.
        Science. 2012; 335: 823-828
        • Strom S.P.
        • Lee H.
        • Das K.
        • Vilain E.
        • Nelson S.F.
        • Grody W.W.
        • Deignan J.L.
        Assessing the necessity of confirmatory testing for exome-sequencing results in a clinical molecular diagnostic laboratory.
        Genet Med. 2014; 16: 510-515
      4. Daly MB, Pilarski R, Axilbund JE, Berry M, Buys SS, Crawford B, et al: National Comprehensive Cancer Network clinical practice guidelines in oncology. Genetic/familial high-risk assessment: breast and ovarian. Version 1.2015. [Internet]. Available at http://www.nccn.org/professionals/physician_gls/pdf/genetics_screening.pdf

        • Shearer A.E.
        • Eppsteiner R.W.
        • Booth K.T.
        • Ephraim S.S.
        • Gurrola 2nd, J.
        • Simpson A.
        • Black-Ziegelbein E.A.
        • Joshi S.
        • Ravi H.
        • Giuffre A.C.
        • Happe S.
        • Hildebrand M.S.
        • Azaiez H.
        • Bayazit Y.A.
        • Erdal M.E.
        • Lopez-Escamez J.A.
        • Gazquez I.
        • Tamayo M.L.
        • Gelvez N.Y.
        • Leal G.L.
        • Jalas C.
        • Ekstein J.
        • Yang T.
        • Usami S.
        • Kahrizi K.
        • Bazazzadegan N.
        • Najmabadi H.
        • Scheetz T.E.
        • Braun T.A.
        • Casavant T.L.
        • LeProust E.M.
        • Smith R.J.
        Utilizing ethnic-specific differences in minor allele frequency to recategorize reported pathogenic deafness variants.
        Am J Hum Genet. 2014; 94: 387-406
        • Spurdle A.B.
        • Healey S.
        • Devereau A.
        • Hogervorst F.B.
        • Monteiro A.N.
        • Nathanson K.L.
        • Radice P.
        • Stoppa-Lyonnet D.
        • Tavtigian S.
        • Wappenschmidt B.
        • Couch F.J.
        • Goldgar D.E.
        • ENIGMA
        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. 2012; 33: 387-406
        • Couch F.J.
        • Rasmussen L.J.
        • Hofstra R.
        • Monteiro A.N.
        • Greenblatt M.S.
        • de Wind N.
        • IARC Unclassified Genetic Variants Working Group
        Assessment of functional effects of unclassified genetic variants.
        Hum Mutat. 2008; 29: 1314-1326
        • Desmond A.
        • Kurian A.W.
        • Gabree M.
        • Mills M.A.
        • Anderson M.J.
        • Kobayashi Y.
        • Horick N.
        • Yang S.
        • Shannon K.M.
        • Tung N.
        • Ford J.M.
        • Lincoln S.E.
        • Ellisen L.W.
        Clinical actionability of multi-gene panel testing for hereditary breast and ovarian cancer risk assessment.
        JAMA Oncol. 2015; ([Epub ahead of print], doi:10.1001/jamaoncol.2015.2690)