Pancreatic ductal adenocarcinoma (PDAC) is a highly aggressive disease with dismal prognosis. Despite the discovery of several promising drug candidates, recent trials with targeted therapies have shown limited or no benefit. For a number of cancers, subclasses have been uncovered that allow the use of therapies that target specific molecular alterations present in only a subset of patients. Recently three distinct subtypes of PDAC have been identified based on gene expression profiles derived from laser-microdissected human tumor tissue. This raised the possibility that differences between such subclasses could be exploited to stratify patients and develop novel targeted therapies. However, a major bottleneck for subtype-specific drug discovery was the lack of pre-clinical models that faithfully recapitulate the full heterogeneity of human pancreatic cancer. The aim of this thesis was the development and verification of such models, analysis of subtype-specific differences and the subsequent discovery of novel treatment options for human PDAC. Our newly developed culture, termed PACO, preserves the molecular features of the corresponding primary tumors, and describes the first models of the exocrine-like subtype. Moreover, orthotopic transplantation of PACO cells re-initiates patient similar tumors that recapitulate these features in vivo, including the characteristic histopathology of the primary tumor specimens. The PACO-cells thus provide a consistent platform for subtype specific in vitro discovery and subsequent in vivo verification. The power of our system is demonstrated by the identification of a novel set of biomarkers that can be easily integrated into routine pathology. Application of this marker set on a tissue microarray representing a cohort of >200 patients revealed that the three subtypes significantly differ in frequency and overall survival.
A combined approach of in silico predictions coupled with biochemical and immunohistochemical verification demonstrated that the three subtypes differ in activation of important oncogenic pathways. Moreover we could show that the three PDAC subtypes vastly differ in their response to cytotoxic and targeted therapy.
Taken together our data demonstrate the need for stratification of PDAC-patients to individualize and improve treatment. Our study introduces novel patient-specific pre-clinical models for pancreatic cancer. Importantly these models are phenotypically consistent from the in vitro to the in vivo application, providing major improvements in pre-clinical drug studies. The PACO model thus enables stratification of individual patients and concomitant personalized treatment according the specific subtypes of human PDAC.
|Supervisor:||Trumpp, Prof. Dr. Andreas|
|Place of Publication:||Deutschland|
|Date of thesis defense:||30 October 2012|
|Date Deposited:||17 Oct 2013 12:09|
|Faculties / Institutes:||The Faculty of Bio Sciences > Dean's Office of the Faculty of Bio Sciences|
|Subjects:||570 Life sciences|