Neural invasion (NI) is encountered in nearly 100% of PDAC patients and is an independent prognostic factor for survival, local recurrence and neuropathic pain. Genetically engineered mouse models (GEMM) of PDAC accurately model both the early and late stages of pancreatic carcinogenesis, but none of these models has ever been reported to mimic all aspects of NI observed in human PDAC.

We have recently characterized the first neuro-invasive PDAC model recapitulating morphologcally neural invasion as observed in human PDAC. In this model ectopic expression of TGFalpha in the pancreas is combined with pancreas specific deletion of p53 and RelA(p65). Departing from this observation we will comprehensively dissect the molecular mechanisms inducing NI. Specifically we will study oncogenic signals, effector pathways and genetic programs mediating NI. We will investigate the temporal dynamics, cellular mechanisms and the transcriptomic features of neural invasion in a large cohort of GEMMs. Furthermore, we are going to discover novel genes and pathways driving neural invasion via a genome-wide PiggyBac transposon-mutagenesis screen and perform in vivo validation and functional characterisation of the identified genes using CRISPR/Cas9-based in vivo gene editing. Moreover, cross-species validation of the identified targets will be performed in human PDAC cohorts and using organoid-neuron co-cultures.

Collectively, this project will study nerve-cancer interactions in PDAC to identify mechanisms underlying this omnipresent, prognostically relevant feature of human PDAC. It will also pursue the long-term goal to provide more human-like, and thus translationally more relevant GEMM of PDAC, and to uncover novel therapeutic targets.