There is obvious merit in being able to accurately predict outcome and tailor treatment according to individual risk and potential for benefit. Epithelial ovarian cancers are characterized by a broad spectrum of biological behavior ranging from tumors that have an excellent prognosis and high likelihood of cure to those that progress rapidly and have a very poor prognosis. This wide clinical spectrum is partly reflected by a number of clinicopathological prognostic variables which include International Federation of Gynecology and Obstetrics stage, histologic subtype and grade, volume of residual tumor remaining after surgical resection, performance status, and age. There has been increasing interest by many groups to incorporate the independent prognostic variables into multivariate models that could better predict outcome. This approach does appear to allow the identification of different prognostic subsets and requires confirmation in prospective studies. There has been, and there continues to be a lot of effort in identifying new prognostic factors that have a biologic rationale and these will be discussed. Most of these new prognostic factors have not been subjected to rigorous testing and this will be clearly necessary before they find clinical application. This is an area that is rapidly evolving with the increased understanding of the molecular basis for ovarian carcinogenesis and progression coupled with technological advances such as DNA arrays and automated polymerase chain reaction. We are at the threshold of developing a new and more objective as well as rational approach to predict prognosis and response to therapy.