Genome-wide regulatory network analysis of high-throughput drug perturbation screens in pancreatic cancer

• Applicants: Michael P. Menden, Ph.D.; Professor Dr. Dieter Saur
• Subject Area: Gastroenterology
• Term: since 2021
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 458890590

Pancreatic ductal adenocarcinoma (PDAC) is a highly aggressive and heterogeneous cancer with a 10-year-survival rate of ~1%. PDAC is remarkably resistant to chemotherapies and almost invariably fatal. A better understanding of the biology and therapeutic vulnerabilities is needed to achieve progress. Our unpublished high-throughput screening for drug sensitivity of PDAC subtypes indicates a great heterogeneity in drug responses, influenced by the genetic, epigenetic and transcriptional identity of each cancer. We performed preliminary transcriptional drug perturbation studies and uncovered a large variety of Mode-of-Actions (MoA), including unique tumour cell transcriptional responses. One example is the induction of the interferon pathway upon treatment, which could significantly boost therapeutic efficacy in vivo. Thus, understanding the molecular responses of PDAC subtypes to drug treatment and analysing the underlying regulatory networks can be a powerful tool to identify new therapeutic approaches and combination strategies.In the proposed research project, we aim at characterising on a large scale the MoA of newly identified therapeutic compounds for the treatment of PDAC in vitro and in vivo. For this purpose, we will first investigate response to 100 drugs by RNA-seq on a comprehensive panel of 200 primary murine cell cultures that cover the full spectrum of PDAC heterogeneity and recapitulate the main genetic alterations observed in patients. Based on this, we will build a ‘context-specific' connectivity map that incorporates the molecular landscape of PDAC and transcriptional perturbation signatures. In addition, we will leverage advanced machine learning and systems biology methods to predict potent drug combinations. The most promising compounds and combinations will then be tested in vivo in fully immunocompetent PDAC subtype models where we will employ single cell RNA-seq to dissect the response of cancer cells and their environment towards drug treatment. This approach allows us to efficiently identify new therapeutic strategies based on tumour cell intrinsic responses, and additional validate these in context of the immune and stromal compartments that are abundant in PDAC. The PDAC in vivo models will mimic closer the expected drug response in patients, enabling hypotheses for combinations with immunotherapies, and ultimately strongly increasing translatability to clinics.

• DFG Programme: Research Grants