Tel.+49 30 838 62450 Fax.+49 30 838 62522
Colorectal cancer (CRC) is the second most common cancer in Germany. 95% of colorectal carcinomas develop via dysplastic changes, so called polyps, in the mucosa of the colon. The limited identification of small polyps and neoplasms with nonpolyploid morphology is a serious problem. During colonoscopy, early dysplastic changes of the mucosa are not recognised in 25-30% of the patients. To improve visualization of surface features of the intestine, detection options need to be optimized. Near infrared (NIR) contrast agents in combination with fluorescence-guided endoscopy are promising approaches to make these precancerous lesions visible. As long as adenoma-specific contrast agents are not available, the technical evaluation of such concepts needs to be developed in animal models and contrast agents for advanced adenocarcinoma.
Orthotopic mouse models of colorectal cancer with cancer cell lines growing in their natural location replicate human disease with high fidelity and are necessary to validate tumour specific contrast agents.
Aim of this dissertation was to develop an orthotopic colorectal cancer model in nude mice with human colorectal cancer cell lines. Tumour growth was monitored by non-invasive bioluminescence imaging as well as by endoscopy. Moreover, the tumour model should be used for NIR imaging with targeted contrast agents.
To replicate human disease with high fidelity three different human colorectal carcinoma cell lines, transfected with a luciferase gene were used. The cells derived from patients with different disease stages and differed in their growth behaviour. The cells were engrafted with different cell suspensions. In different cell suspension groups different take rates were observed.
By doubling the number of injected cells, the take rate was improved. Especially by the use of Matrigel, a substance which contains growth factors for cells, high take rates were observed. All three cell lines were implanted successfully into the colon wall of nude mice, differences in take rates between different cell lines were not observed.
Histological and immunohistochemical stainings confirmed the localization of primary tumours in the colon wall and the human origin.
Tumour growth and metastasis were monitored in vivo using bioluminescence by the stable luciferase expression of the transplanted cells and luciferin injection in mice.
In addition tumour growth was observed directly using a small animal endoscope.
Integrins play an important role in the development and growth of colorectal tumours. First NIR studies on tumor-bearing mice were made using a specific contrast agent, called IntegriSense680. IntegriSense680 is an integrin antagonist for detection of αvβ3 integrin and is coupled to a NIR dye. The contrast agent was injected in tumor-bearing animals. After euthanasia the intestine was examined for the distribution of contrast agent with an NIR laser fiber endoscope. Using this contrast agent we could clearly distinguish between normal mucosa and tumour tissue. Colocalization of fluorescence and bioluminescence proofed that the contrast agents was in bioluminescence positive areas. These results were also confirmed by imaging tumour cryosections with a NIR scanner.
The newly established orthotopic CRC tumour model now provides the basis for further pre-clinical studies to validate new contrast agents and new targets for the diagnosis and therapy of colorectal cancer. As soon as adenoma-specific biomarkers are available, the results of fluorescence and bioluminescence imaging can be used as a basic concept for further research. The combination of technical and molecular results will hopefully lead to the development of fluorescencebased contrast agents for preventive colonoscopy in patients.