Following breast and lung cancer, colorectal cancer is the third most common cancer seen in the UK. In Z008, 39,991 cases of large bowel cancer were registered, and colorectal cancer remains the second most common cause of death from cancer behind lung cancer. Carcinoembryonic antigen (CEA) was first identified in 1965 in human colon cancer tissue extracts. Since then it has been found being expressed in other cancer patients, such as those with pancreatic, lung and breast carcinoma. Although it has been identified in other cancer types, it is for its role in gastric cancer that we are currently investigating the use of this antigen for a novel targeted chemotherapeutic method known as Antibody-Directed Enzyme Prodrug Therapy (ADEPT). ADEPT is a technique that delivers a cytotoxic agent specifically to tumour sites. An antibodyenzyme complex (fusion protein) is administered to the patient and after allowing sufficient time a clearing agent is added to remove any unbound fusion protein. A prodrug is then administered and in the presence of the fusion protein, the prodrug becomes a cytotoxic agent able to kill tumour cells in the vicinity of the tumour giving more specificity of action than traditional cytotoxic agents. ADEPT has been trialled clinically before with mixed results, however this project is for an "enhanced-ADEPT" system, an idea that has received EUROSTARS funding. As with previous versions of ADEPT ScFv (Single chain variable Fragment) fusion protein fusion proteins are to be used. These fusion proteins have had key features re-designed by molecular biological methods in order to provide step changes in performance over previous ADEPT systems. To generate the fusion proteins we are investigated, we used the yeast expression system Pichia pink. The yeast expression host is an ideal tool to manufacture fusion proteins. The Pichia pink expression system is capable of post-translational modification of proteins and can secrete completed fusion proteins into the growth media, allowing for a simple method of harvesting and purification. The Pichia pink system also gives some control over glycosylation and limiting proteolysis. The aims of the project were to generate and optimise expression of anti CEA fusion protein using the Pichia pink system; and characterise anti CEA fusion protein, using a variety of experimental techniques including flow cytometry, confocal microscopy, biolayer interferometry and cytotoxicity assays. Two novel ScFv fusion protein's were expressed in Pichia pink strain 1. Different production batches of these were shown to have variable enzyme activity as determined by a methotrexate hydrolysis assay, between 0 and 115U jmL. Specific binding of these fusion proteins to CEA was observed using biolayer interferometry, with average KD values in the low nanomolar range. While no binding of HEKZ93, a CEA negative cell line was observed, fusion proteins were found to adhere to both LoVo and MKN45 CEA positive cell lines, via flow cytometry and confocal microscopy. The fusion proteins were shown, in the presence of prodrug to cause the death of MKN45 cells in a modified clonogenic assay, while HEKZ93 cells remained largely unaffected. The CEA positive cell line LoVo however exhibited resistance to the drug. Further testing of these fusion proteins is certainly recommended.