Over the past two decades, empirical optimization of cytotoxic chemotherapy combinations and surgical debulking procedures have improved outcomes and survival in epithelial ovarian cancer. Yet, this disease remains the fifth leading cause of cancer-related deaths in the US, as cure rates seem to have reached a plateau at approximately 20% with conventional chemotherapy. Novel high-throughput genomic and proteomic analyses have improved the molecular understanding of ovarian carcinogenesis, thereby providing a vast array of new potential drug targets with complex signalling interactions. In order to yield the most significant impact on disease outcome, it is necessary to carefully select, and subsequently target, the driving molecular pathway(s) within a tumour or tumour subtype, which are most likely to correspond to high-frequency mutations and genomic aberrations. The identification of biomarkers predictive of response to targeted therapy is essential to avoid poor responses to potentially useful drugs in unselected trial populations. With some promising, albeit early, phase III data on the angiogenesis inhibitor bevacizumab, exciting new opportunities lie ahead with the ultimate goal of personalizing therapies to individual tumour profiles.