Abstract Background Preclinical evaluations of chemotherapies depend on clinically relevant animal models for pancreatic cancer. The injection of syngeneic murine adenocarcinoma cells is one efficient option to generate carcinomas in mice with an intact immune system. However, this option is constrained by the paucity of appropriate cell lines. Results The murine pancreatic adenocarcinoma cell lines 6606PDA and 7265PDA were compared to the 6606l cell line isolated from a liver metastasis from mice suffering from pancreatic cancer. In tissue culture 6606PDA and 6606l proliferated faster than 7265PDA. 7265PDA cells were, however, significantly more sensitive to gemcitabine as assessed by BrdU-incorporation and trypan blue exclusion assays in vitro. Within 1 week after injection of either one of these three cell lines into the pancreas of C57BL/6J mice, carcinomas were observed by T2 weighted magnetic resonance imaging and histology. Three weeks after injecting 6606PDA or 6606l cells large carcinomas could be characterized, which were surrounded by extensive desmoplastic reaction. After injection of 7265PDA cells, however, remission of cancer was observed between the first and the third week. Compared to 6606PDA cell derived carcinomas a higher apparent diffusion coefficient was quantified by diffusion weighted magnetic resonance imaging in these tumors. This correlated with reduced cancer cell density observed on histological sections. Conclusion All three cell lines can be used in vitro for testing combinatorial therapies with gemcitabine. The 6606PDA and 6606l cell lines but not the 7265PDA cell line can be used for evaluating distinct therapies in a syngeneic carcinoma model using C57BL/6J mice. Diffusion-weighted MRI proved to be an appropriate method to predict tumor remission.