AccuratePredictionof BRCA1 and BRCA2 HeterozygousGenotype UsingExpressionProfilingafterInducedDNADamage ZsofiaKote-Jarai, 1 Lucy Matthews, 2 Ana Osorio, 1,4 SusanShanley, 1,5 IanGiddings, 2 FrancoisMoreews, 2 ImogenLocke, 1,5 D.GarethEvans, 6 DianaEccles, 7 TheCarrierClinicCollaborators, 5 RichardD.Williams, 3 MarkGirolami, 8 Colin Campbell, 9 andRosEeles 1,5 Abstract Purpose: In this study, the differential gene expression changes following radiation-induced DNA damage in healthy cells from BRCA1/BRCA1 mutation carriers have been compared with controls usinghigh-density microarray technology.We aimed to establishif BRCA1/BRCA2 mutationcarrierscouldbedistinguishedfromnoncarriersbasedonexpressionprofilingofnormal cells. Experimental Design: Short-term primary fibroblast cultures were established from skin biopsies from10 BRCA1 and10 BRCA2 mutationcarriersand10controls,allof whomhadprevi- ously had breast cancer.The cells were subjected to 15 Gy ionizing irradiation to induce DNA damage. RNAwas extracted from all cell cultures, preirradiation and at1hour postirradiation. For expression profiling,15 K spotted cDNA microarrays manufactured by the Cancer Research UK DNA Microarray Facility were used. Statistical feature selection was used with a support vector machine (SVM) classifier to determine the best feature set for predicting BRCA1 or BRCA2 heterozygous genotype.To investigate prediction accuracy, a nonprobabilistic classifier (SVM) andaprobabilisticGaussianprocess classifier wereused. Results: In the taskof distinguishing BRCA1 and BRCA2 mutation carriers fromnoncarriers and from each other following radiation-induced DNA damage, the SVM achieved 90%, and the Gaussian process classifier achieved100% accuracy.This effect could not be achieved without irradiation. In addition, the SVMidentified a set of BRCA genotypepredictorgenes. Conclusions: We conclude that after irradiation-induced DNA damage, BRCA1 and BRCA2 mutation carrier cells have a distinctive expression phenotype, and this may have a future role in predicting genotypes, withapplication to clinical detectionand classificationof mutations. It is estimated that 5% to 10 % of breast cancer patients develop the disease due to the presence of a mutation in a breast cancer predisposition gene (1). A significant proportion of this population (about 50%) has a mutation in one of the known breast cancer predisposition genes (BRCA1 or BRCA2 ). Besides the definite disease-causing deleterious mutations, small alter- ations, such as single base substitutions (missense mutations), are frequently found in these genes. Their functional effects are usually unknown, hence they are termed variants of uncertain significance. Some of these variants of uncertain significance could also have a role in breast cancer predisposition, but it is not currently possible to establish their disease-causing effect. The available diagnostic tests for mutation analysis of BRCA1/ BRCA2 are time and labor intensive, expensive, and do not allow for the identification of all types of mutation. The aim of this study was to determine whether gene expression profiling could be used to distinguish between heterozygous BRCA1 and BRCA2 mutation carriers and control samples. Human Cancer Biology Authors’Affiliations: 1 Translational Cancer GeneticsTeam and Sections of 2 Molecular Carcinogenesis and 3 Paediatric Oncology,The Institute of Cancer Research, Sutton, Surrey, United Kingdom; 4 Department of Human Genetics, Spanish National Cancer Centre, Madrid, Spain; 5 Royal Marsden NHSFoundation Trust,London,UnitedKingdom; 6 St.Mary’sHospital,Manchester,UnitedKingdom; 7 Clinical Genetics, Princess Anne Hospital, Southampton, United Kingdom; 8 Department of Computing Science, Bioinformatics Research Centre, University of Glasgow, Glasgow, United Kingdom; and 9 Computational Intelligence Group, Universityof Bristol, Bristol, United Kingdom Received12/27/05;revised2/7/06;accepted3/23/06. Grant support: Medical Research Council Discipline Hopping Award (M. Girolami), Cancer Research UK (L. Matthews, I. Giddings, F. Moreews, and the microarray production), NHMRC Australia (S. Shanley), Cancer Research UK grants C5047/A5463 (I. Locke) and C5047/A3354D (The Carrier Clinic and R. Eeles), Engineering and Physical Sciences Research Council grant GR/R96255/01 (C. Cambell), Breast Cancer Campaign project grant (Z. Kote-Jarai and the study), legacy of the late Marion Silcock (Z. Kote-Jarai and the study), and Maxse/Knowles Research Fund (Z. Kote-Jarai and the study). Thecostsofpublicationofthisarticleweredefrayedinpartbythepaymentofpage charges.This article must therefore be hereby marked advertisement inaccordance with18U.S.C.Section1734solelytoindicatethisfact. Note: Supplementary data for this article are available at Clinical Cancer Research Online(http://clincancerres.aacrjournals.org/). A. Osorio was a Haddow Fellow ofThe Institute of Cancer Research. D. Gareth Evans,D.Eccles,andR.Williamshadnospecificfundingrelatedtothiswork. The Carrier Clinic Collaborators areAudreyArdern-Jones, Elizabeth Bancroft, Kate Bishop, Elly Lynch, Rebecca Doherty, SarahThomas, Asher Salmon, Clare Turnbull, SameerJhavar. Requests for reprints: Zsofia Kote-Jarai,Translational Cancer ResearchTeam, The Institute of Cancer Research,15 Cotswold Road, Sutton, Surrey, SM2 5NG United Kingdom. Phone: 44-208-661-3105; E-mail: zsofia.kote-jarai@icr.ac.uk. F 2006AmericanAssociationforCancerResearch. doi:10.1158/1078-0432.CCR-05-2805 www.aacrjournals.org ClinCancerRes2006;12(13)July1,2006 3896 Cancer Research. on January 10, 2022. © 2006 American Association for clincancerres.aacrjournals.org Downloaded from