Animal models of cognate human conditions permit the rigorous exploration of mechanisms of disease pathogenesis and provide systems to devise and test therapeutic and detection strategies. Although inbred mouse strains offer many advantages over other animals for investigations of malignant disease, mice unfortunately do not develop PDA spontaneously or even following carcinogen treatment. Recent technological developments in conditional gene targeting have enabled the generation of mutant mouse cancer models of PanIN and PDA that closely mimic human pancreatic cancer and are thus suitable for investigations of the human disease process. In this paper, we summarize the knowledge of the human disease that has informed the ability to generate accurate mouse models of pancreatic cancer and describe potential biological and clinical applications.
EPIDEMIOLOGY OF PANCREATIC CANCEREpithelial carcinomas account for the majority of cancer deaths in the United States, and among them, pancreatic cancer is the fourth most common with the highest case-fatality ratio. Pancreatic ductal adenocarcinomas (PDA) comprise more than 90% of pancreatic cancer cases and are its most lethal form among an assortment of additional histological subtypes including cystic neoplasms, acinar carcinomas, and islet cell endocrine tumors. Risk factors associated with the development of PDA include increased age, tobacco usage (Silverman et al. 1994), African heritage (Ahlgren 1996), and a family history of pancreatic cancer . Despite the relatively large numbers of patients stricken with PDA, estimated at 32,180 for 2005, our expanding knowledge of critical aspects of this disease remains terribly inadequate and undoubtedly contributes to our inability to intervene meaningfully. The impetus to investigate the biological underpinnings of PDA is further motivated by the predicted doubling in PDA incidence over the next two decades as the population ages (Hruban et al. 2006a). Nonetheless, the public investment in pancreatic cancer research is woefully inadequate relative to other malignancies, and consequently, the number of investigators pursuing PDA is unfortunately restricted.
GENETICS OF PDA AND PanINPDA is thought to evolve from a preinvasive precursor state termed pancreatic intraepithelial neoplasms (PanINs) because PanINs demonstrate cytological and genetic changes that reflect those evident in invasive PDA. PanINs are divided into three stages characterized by increasing degrees of cellular and architectural atypia. PanIN-1A lesions are flat and contain tall columnar ductal cells in distinction to the short cuboidal cells that line normal pancreatic ducts; PanIN-1B lesions additionally demonstrate mild papillary architecture in the ducts but no nuclear atypia; the PanIN-2 stage signifies cells with moderate nuclear atypia and loss of cellular polarity; PanIN-3s represent the carcinoma-in-situ stage of pancreatic cancer, characterized by marked atypia, dysregulated growth, cribriform structures, and pinched-off clusters of epithelial cells...