The treatment of epithelial ovarian cancer (EOC) is narrowly focused despite the heterogeneity of this disease in which outcomes remain poor. To stratify EOC patients for targeted therapy, we developed an approach integrating expression and genomic analyses including the BRCAness status. Gene expression and genomic profiling were used to identify genes recurrently (>5%) amplified and overexpressed in 105 EOC. The LST (Large-scale State Transition) genomic signature of BRCAness was applied to define molecular subgroups of EOC. Amplified/overexpressed genes clustered mainly in 3q, 8q, 19p and 19q. These changes were generally found mutually exclusive. In the 85 patients for which the genomic signature could be determined, genomic BRCAness was found in 52 cases (61.1%) and non-BRCAness in 33 (38.8%). A striking mutual exclusivity was observed between BRCAness and amplification/overexpression data. Whereas 3q and 8q alterations were preferentially observed in BRCAness EOC, most alterations on chromosome 19 were in non-BRCAness cases. CCNE1 (19q12) and BRD4 (19p13.1) amplification/overexpression was found in 19/33 (57.5%) of non-BRCAness cases. Such disequilibrium was also found in the TCGA EOC data set used for validation. Potential target genes are frequently amplified/overexpressed in non-BRCAness EOC. We report that BRD4, already identified as a target in several tumor models, is a new potential target in high grade non-BRCAness ovarian carcinoma.Epithelial ovarian cancer (EOC) is the seventh most common cause of cancer related death in women. 1 This high mortality rate is a result of delayed diagnosis and limited therapeutic options. The reference treatment of EOC is an association of surgical cytoreduction and platinum-based chemotherapy. About 80% of tumors respond to the first line of treatment 2 but a relapse develops in most cases and there is no curative treatment for recurrent diseases.EOC spans many diseases that differ in their histology, primary anatomical location and molecular features. Histologically, serous (40%), endometrioid (24%) mucinous (6%) and clear cells (26%) are the most frequent. 3 The classical three-tier histological grading system was recently modified as a binary system for serous carcinoma. 4 Anatomically, though a subset of endometrioid and clear cell carcinoma are known to develop from pelvic endometriosis, the extensive study of prophylactic oophoro-salpingectomy specimens revealed that a notable proportion of high grade serous ovarian cancers are actually derived from the distal fallopian tube. 5 Bio-pathological studies have shown that histological subtypes progress through distinct biological pathways. Two