Epithelial ovarian cancer (EOC) is the leading cause of death from gynaecological malignancies in the UK. Currently, serum CA125 level is the only circulating cancer biomarker routinely used in clinical practice to monitor cancer recurrence and response. However, it has low sensitivity and specificity for diagnosis. Biomarkers for screening, diagnosis, prognosis and outcomes prediction are lacking. There is a clear need to identify such biomarkers to assist clinical decision making and help to elucidate cancer- associated biological processes. The aim of this project was to evaluate the diagnostic, prognostic and predictive value (both surgery and chemotherapy) of circulating proteins in EOC. Data on 41 circulating proteins generated on a multiplex Luminex ELISA platform using plasma was used for biomarker discovery. Subsequent validations were performed using ELISAs, as well as publicly available microarray datasets. This body of work constitutes the following observations: 1) The combination of CA125 and HE4 are confirmed to be the current best performing biomarker for distinguishing EOC from non-EOC; 2) FOLR1 is a potential biomarker for differential diagnosis between borderline tumours of ovary and EOC; 3) Circulating levels of PAI1 active and SDC1 are predictive biomarkers for surgical outcome, and indicate that the biology of tumour-stromal interaction plays a key role in tumour resectability; 4) Circulating SDC1 is also a poor prognostic factor associated with thrombocytosis and PDGF signalling, and low SDC1 levels may predict benefit from anti-angiogenics such as bevacizumab; and 5) High circulating FOLR1 levels are associated with poor prognosis, and may represent a biomarker for patient selection for FOLR1- targetted therapies in development.