Transcriptional characteristics of familial non-BRCA1/BRCA2 breast tumors Ricardo Ferna ´ndez-Ramires 1 , Gonzalo Go ´mez 2 , Iva ´n Mun ˜oz-Repeto 1 , Loris de Cecco 3,4 , Gemma Llort 5 , Alicia Cazorla 6 , Ignacio Blanco 5 , Manuela Gariboldi 3,4 , Marco Alessandro Pierotti 3,4 , Javier Benı ´tez 1 and Ana Osorio 1 1 Human Genetics Group, Human Cancer Genetics Program, Spanish National Cancer Center (CNIO) and CIBERER, Madrid, Spain 2 Bioinformatics Unit (UBio), Structural Biology and Biocomputing Program, Spanish National Cancer Center (CNIO), Madrid, Spain 3 Fondazione Istituto Nazionale dei Tumori, Milan, Italy 4 Fondazione Istituto FIRC Oncologia Molecolare, Milan, Milano, Italy 5 Genetic Counseling Unit, Catalan Institute of Oncology, IDIBELL, L’Hospitalet, Barcelona, Spain 6 Department of Pathology, Fundacio ´n Jime ´nez Dı ´az, Madrid, Spain To better understand the alterations present in the group of the so-called BRCAX tumors, we have used a cDNA microarray containing genes related to tumorigenesis and analyzed a series of 49 tumors consisting of 13 BRCA1, 14 BRCAX and 22 sporadic. We have confirmed that the BRCAX tumors are heterogeneous and can be divided in at least two main subgroups, so-called A and B, transcriptionally distinguishable and with different altered pathways within each of the groups. We have found that BRCAX-A and B subgroups, can be classified as Luminal A and Luminal B, respectively, taking into account the intrinsic phenotypes defined for the sporadic breast tumors. We have found that, at the somatic level, the BRCAX-B tumors are identical to their sporadic Luminal B counterparts, whereas BRCAX-A, despite having a Luminal A phenotype, shows additional genomic alterations. We have found 21 deregulated genes in the BRCAX-A group that we have called ‘‘the BRCAX susceptibility pathway’’ and suggested it as a candidate to search for new genes involved in the inherited susceptibility underlying the disease in this group. Breast cancer is the most common neoplasia among women and it is estimated that 5–10% of the cases present a familial history of the disease. Germ-line mutations in the high-sus- ceptibility genes BRCA1 and BRCA2 are responsible for around 25% of hereditary breast and ovarian cancer syndrome cases, 1–3 but the etiology of the remaining familial breast can- cer cases (so-called BRCAX cases) remains unclear. Mutations in genes such as TP53 and PTEN have been linked with high risk for breast cancer within specific cancer syndromes, 4,5 and germ-line variants in other genes such as CHEK2 and ATM, 6 and more recently, the Fanconi Anemia genes BRIP1/FANCJ and PALB2/FANCN 7,8 have been found to confer modest risk to familial breast cancer. However, alterations in these addi- tional genes account for less than 5% of the hereditary cases and are sometimes restricted to specific populations. Various linkage analyses performed in BRCAX families have pointed to different candidate regions, 9–13 but to date none of them have been confirmed to actually contain a new high-pene- trance breast cancer susceptibility gene. A major limitation of this type of studies seems to be the genetic background de- pendence as detected, for example, in genome-wide linkage scans performed in Swedish (10q23.32-q25.3), 9 Euro-Austra- lian (2p), 12 Scandinavian (13q21-q22) 14 and Spanish (3q26, 6q25 and 21q22) 11 populations. The failure to find new high- penetrance susceptibility genes, together with the discovery of moderate risk variants, has progressively supported the idea of the polygenic model, that is, the residual genetic variance is explained by a large number of variants conferring a moderate effect on breast cancer risk. 15 The scene has recently taken a turn with the report of high-penetrance mutations in the new Fanconi Anemia gene, RAD51C in German breast and ovarian cancer families not associated with mutations in BRCA1 and BRCA2 16 If this finding is confirmed we could be dealing with the third high-risk breast cancer gene in years but again, this would explain less that 1% of the high-risk families which shows once again that the likelihood to find new susceptibility genes is limited by the great genetic heterogeneity. Key words: expression profiling, familial breast cancer, BRCAX, susceptibility genes Additional Supporting Information may be found in the online version of this article. Grant sponsor: Spanish Health Ministry FIS; Grant number: PI061090; Grant sponsors: Biomedical Network Research Centre for Rare Diseases (CIBERER), Marie Curie Sixth Framework, Caja Navarra, Spanish Ministry of Health, Asociacio ´n Espan ˜ola Contra el Ca ´ncer (AECC) DOI: 10.1002/ijc.25603 History: Received 11 May 2010; Accepted 23 Jul 2010; Online 16 Aug 2010 Correspondence to: Ana Osorio, Human Genetics Group, Human Cancer Genetics Program, Spanish National Cancer Center (CNIO) and CIBERER, Melchor Ferna ´ndez Almagro 3, Madrid 28029, Spain, E-mail: aosorio@cnio.es Cancer Genetics Int. J. Cancer: 128, 2635–2644 (2011) V C 2010 UICC International Journal of Cancer IJC