Resectable transgenic tumor models in mice for biliary cancers

Surgical complete resection (R0) of the primary tumor is the only potential curative treatment option for many tumor entities. Despite surgical resection, patients with intrahepatic cholangiocarcinoma (ICC) or pancreatic ductal adenocarcinoma (PDAC) have poor prognosis, because of frequent tumor recurrence and outgrowth of metastases after surgery. Unfortunately, the current animal tumor models do not reflect the aspect of resectability, though surgery and adjuvant therapy is still the most important and potentially curative therapeutic intervention for cancer patients.

Aim: To address the aspect of tumor resection in murine tumor models, we established a corresponding model for ICC or PDAC in mice by facilitating endogenous induction of a single tumor in situ that is suitable for surgical resection.

Methods: Intratissue injection of Sleeping Beauty-based, oncogenic transposon plasmids followed by subsequent local electroporation leads to formation of a locally restricted, resectable primary tumor. The injected DNA includes a transposon vector for expression of mutated KRas-G12V and a plasmid for expression of Cre-recombinase. In p53-fl/fl-mice, the Cre recombinase induces the knockout of p53 and simultaneous transformation by genomic insertion and expression of the RasG12V transposon.

Results: According to our aims, mice developed a single tumor lesion at the electroporated tissue locus. Formation of ICC or PDAC was verified by histological analysis. Molecular analysis after electroporation provided evidence for acinus cells as origin of tumor formation. Metastases in the lung, liver and peritoneum could be detected. By R0-resection of the primary tumor, we were able to prolong median survival with the observation of local disease recurrence, peritoneal carcinomatosis, and metastases in liver and lung. Using gemcitabine as therapeutic standard for biliary cancer, we could observe a survival benefit only in the adjuvant approach. Palliative gemcitabine application did not improve survival. These results indicate different mode of action of gemcitabine in the adjuvant or palliative situation, respectively. In the palliative setting with gemcitabine the myeloid suppressor cell population was essentially reduced. However, an increasing population of intratumoral regulatory T cells maintained the immune suppressive tumor microenvironment explaining the lacking efficacy of gemcitabine in the palliative situation.

Conclusion: Our resection models reflect the clinical situation in humans and holds great promise for preclinical evaluation of novel multimodal and adjuvant therapies in genetically defined biliary cancers to prevent recurrence and outgrowth of metastases. Furthermore, the results suggest that therapies have to be evaluated separately for palliative or adjuvant approaches, respectively.