Background:Pancreatic cancer is the most lethal common cancer and fifth leading cause of cancer mortality in Australia. The prognosis for those diagnosed is extremely poor, with only 2 – 9% of patients surviving five years after diagnosis. Pancreatic tumours are often highly aggressive and cause ambiguous symptoms, making early detection of the disease extremely difficult. Additionally, current blood biomarkers lack the sensitivity and specificity to be used for non-invasive screening. Expression of N-glycolylneuraminic acid (Neu5Gc)-containing glycoconjugates is a significant component of the aberrant glycosylation observed in human tumour cells. As such, Neu5Gc has been proposed as a cancer biomarker. Through purpose engineering of the B subunit of the subtilase cytotoxin (SubB), the resulting SubB2M lectin binds Neu5Gc with high specificity, and has been used in surface plasmon resonance (SPR) assays to measure Neu5Gc levels in the serum of ovarian cancer patients, demonstrating elevated levels at all stages compared to cancer-free controls. Neu5Gc has not previously been measured in the serum of pancreatic cancer patients, though it has been identified in pancreatic cancer tissue. As pancreatic tumour cells are known to upregulate sialyltransferases and secrete sialoglycoproteins, there is great potential for Neu5Gc to be used as a serum biomarker for pancreatic cancer. Furthermore, cancer antigen 19-9 (CA 19-9, also known as sialyl Lewis a), the best validated clinical biomarker for pancreatic cancer, possesses a sialic acid and is expressed on large glycoproteins such as MUC1, along with other pancreatic cancer antigens, suggesting Neu5Gc may be used to improve the accuracy of current biomarkers. The overall objective of this study was to assess the levels of Neu5Gc biomarkers in sera from people with pancreatic cancer. Methods: Serum samples from female (n=42) and male (n=17) patients with pancreatic cancer (stages I – IV) were obtained from the Victoria Cancer Biobank (VCB), along with cancer-free female (n=12) and male (n=12) samples. The serum Neu5Gc levels of the samples was measured by surface plasmon resonance (SPR), utilising the SubB2M lectin with appropriate controls and standards. The SubB2M lectin was expressed and purified as a recombinant protein. Improvement of SubB2M protein yields was attempted through modification of the purification protocol, and addition of a C-terminal penta-Lysine tag. An anti-CA 19-9/SubB2M dual-factor ELISA was developed for the purpose of detecting Neu5Gc-decorated CA 19-9. The technique used anti-CA 19-9 monoclonal antibodies (mAbs) for the capture of CA 19-9 protein from samples, and biotin-conjugated SubB2M for detection of Neu5Gc on captured material. Three serum samples were measured with this method, with the aim of distinguishing pancreatic cancer patient serum samples from cancer-free samples when both possess high CA 19-9 serum levels.Results: Serum Neu5Gc levels were found to be significantly elevated at all stages of pancreatic cancer, with stages I (p ≤ 0.0001) and II (p ≤ 0.0001) displaying the highest average levels compared to other stages. Overall, Neu5Gc had an area under the receiver operating characteristic (ROC) curve (AUC) of 0.8962 (95% CI 0.8224 – 0.0.9699) for the entire cohort, and an AUC of 0.9762 (95% CI 0.9301 – 1.000) and 0.9755 (95% CI 0.9304 – 1.000) for females and males respectively. No difference was found in serum Neu5Gc levels between patients with ductal and neuroendocrine tumours. SubB2M protein could be purified using the standard protocol, however, its yield was enhanced by inclusion of 0.05% TWEEN-20 to buffers during concentration and buffer-exchange procedures. Addition of the penta-Lysine tag resulted in loss of Neu5Gc recognition. The anti-CA 19-9/SubB2M dual-factor ELISA could detect CA 19-9 control protein standards in a concentration dependent manner, however the assay failed to discriminate elevated CA 19-9 evels in a cancer patient from a cancer-free subject. Conclusion: Using the SubB2M lectin, elevated levels of Neu5Gc could be detected in the serum of patients with all stages and types of pancreatic cancer. The SubB2M-A12-SPR assay can predict the presence of pancreatic cancer in a cohort of pancreatic cancer patients and cancer-free controls with an accuracy greater than CA 19-9, suggesting these findings are of clinical significance. We investigated whether Neu5Gc is expressed on the CA 19-9 antigen. Though the anti-CA 19-9/SubB2M dual-factor ELISA failed to distinguish a cancer patient from a cancer-free individual with high CA 19-9 levels, the potential for Neu5Gc as a biomarker in pancreatic cancer is evident, and further investigation using the SubB2M lectin will lead to improvements in diagnostic methods.