Hoffmann, PeterOehler, MartinRicciardelli, CarmelaMartin, Karina2016-03-312016-03-312015http://hdl.handle.net/2440/98148Epithelial ovarian cancer accounts for 5% of all cancer deaths and greater than 50% of all gynaecological cancer deaths. It presents at a late clinical stage in more than 60% of patients, and is associated with a 5-year survival of only 30% in this group. In contrast, the 5-year survival for patients with organ-confined stage I ovarian cancer exceeds 90%, and most patients are cured of their disease. Thus, the detection of early stage ovarian cancer is the best way to improve survival. No clinically applicable method exists for the early detection of ovarian cancer. Hence, there is an unmet medical need for an accurate screening test. Most scientific efforts towards early detection are focused on the discovery of tumour-associated antigens (TAA). Autologous antibodies against TAAs, however, may serve as more sensitive diagnostic markers. They circulate in the blood before TAAs and are usually more abundant than the TAAs themselves as a result of amplification through the humoral immune response. Accumulating evidence also suggests that a humoral response already exists during malignant transformation when aberrant gene expression is translated into premalignant cellular changes. In this thesis, potential autoantibody biomarkers for ovarian cancer were discovered, verified and validated as an early detection test. A new immunoproteomic strategy was developed to identify novel autoantibodies that were elevated in serous ovarian cancer patients. Lysate extracted from the ovarian tissue of a patient was applied to an immunoaffinity column generated with autologous antibodies and a paired control immunoaffinity column. Relative quantification of captured autoantigens was performed using isotope coded protein label (ICPL) technology coupled with high resolution LC-MS. At a protein ratio cut-off of 1.45-fold, 148 autoantibodies were found to be enriched in ovarian cancer patients compared to the corresponding controls. Upon bioinformatic prioritisation 50 autoantibody candidates were selected for verification. Protein microarray analysis of 98 samples revealed 9 autoantibody candidates to be significantly different in early stage cancer patients compared to healthy and benign controls. Biomarker candidates anti-ANXA1, anti-SAHH and anti-ARP3 showed the greatest potential where each marker achieved greater than 90% specificity at 83.3% sensitivity. As a 4-biomarker panel with the ‘gold standard’ for ovarian cancer detection, cancer antigen (CA)125, a sensitivity of 76.5% at 100% specificity was attained. These values of sensitivity and specificity for early stage ovarian cancer surpassed the minimum requirements for an implementable screening test and showed great promise as a diagnostic tool. Validation of the top three autoantibody candidates using protein microarray revealed anti-ANXA1 to be the most robust and effective biomarker for stage I cancer detection. As a single biomarker anti-ANXA1 had 81.8% sensitivity and 71.9% specificity for stage I cancer compared to healthy and benign controls. Excitingly, in combination with CA125, a sensitivity of 71.4% at 100% specificity was achieved when differentiating stage I cancer patients from healthy individuals. For this level of effectiveness a positive and negative predictive value of 100% and 99.99% was achieved, respectively. Therefore, a biomarker panel containing anti-ANXA1 and CA125 may enable the development of a detection test that can be used to screen for stage I serous ovarian cancer in the general population. This promising discovery demands further investigation where continuing analysis in prospective samples is essential. The discovery of a screening test is crucial to reduce the morbidity and mortality caused by ovarian cancer. This study investigates the presence of differential autoantibody signatures in serous ovarian cancer patients as potential biomarkers. Additionally, those identified TAAs that are functionally involved in carcinogenesis could also serve as therapeutic targets. Finally, the immunoproteomic approach developed here could be used in studies aiming to discover novel autoantibody biomarkers for other asymptomatic malignancies.biomarkersovarian cancerautoantibodiesimmunoproteomicsmass spectrometryTheses