This research project exploits the design flexibility of recombinant antibodies (rAbs) coupled with selective biomarkers to improve diagnosis and treatment approaches of the most common pancreatic cancer, known as pancreatic ductal adenocarcinoma (PDAC). PDAC is a malignancy with almost a 100% mortality rate, as a result of resistance to chemotherapy and very inadequate diagnostic methods. A panel of recombinant scFv antibody fragments, specific for the selected biomarkers, anterior gradient 2 (AGR2), mesothelin (MSLN), and mucin 1 (MUC1) were generated for use to improve the diagnosis of PDAC. Additionally, a bispecific T-cell engager was created for potential targeted treatment. Initially, a murine-derived Capan-1 specific scFv library with a serum antibody titre of 1/200,000 was generated, following immunisation with the pancreatic cancer cell line, Capan-1, which expressed the biomarkers of interest. Library screening was carried out using MSLN expressed in HEK293 cells, following transfection with a MSLN-containing plasmid, and a MSLN-specific scFv was successfully isolated. The resulting scFv were characterised as a immunohistochemistry-based antibodies for the detection of MSLN in tumour tissue and found to have a working concentration of 1/500 when analysed by western and dot blots. The human naïve library, Tomlinson I and J, was employed for the isolation of scFv specific to MUC1 and AGR2. Functional specific scFv for both antigens were found through panning with purified MUC1 and AGR2 proteins. A CD3ɛ-specific scFv with a limit of detection of 20 pg/mL was isolated from an available library derived from an appropriately immunised avian host. A bispecific T-cell engager (BiTE) was constructed using the anti-CD3ɛ scFv and a previously generated anti-HER2 scFv with the ability to detect down to 10pg/mL of HER2 in human serum. Two proof-of-concept methods of BiTE construction, employing PCR and cloning techniques are described. The binding abilities of the BiTE to each antigen individual and in dual format was evaluated. The generated BiTE successfully showed the ability to bind to both antigens simultaneously in ELISA format .