The ability of selected cancers to invade in, around, and along nerves is a well described event termed perineural invasion (PNI). PNI has repeatedly been shown to be an adverse pathologic finding, heralding more aggressive local disease, higher recurrence rates, and worsened survival. Despite widespread recognition of the clinical significance of PNI, its mechanisms remain poorly understood. We recently demonstrated the critical role of cancer cell chemotaxis (external gradient sensing and directional migration) towards glial-derived neurotrophic factor (GDNF) and activation of the RET tyrosine kinase receptor as a key mechanism for PNI. Mechanisms of chemotaxis have been previously defined primarily in the non-malignant systems of dictyostelium discoideum (slime mold) and leukocytes. However, chemotactic mechanisms have not been as well defined in cancer cells. Interestingly, pathways associated with chemotaxis in dictyostelium and leukocytes are frequently aberrant in cancer. These include tyrosine kinase receptor activity, PI3K activity, Rho family (Cdc42, Rac1, RhoA) GTPase and Ras GTPase activity. GDNF has been shown to activate each of these factors in pancreatic cancer cells. We will elucidate the relationship between oncogenic activation of these factors and their chemotactic response to GDNF, in the clinically important context of PNI. Our central hypothesis is that the dysregulation of these factors in some cancers may promote chemotaxis towards GDNF, leading to PNI. We will define the oncogenic characteristics that promote a neural invasive phenotype. We will define the roles of these pathways in PNI using a combination of techniques including: single cell live time-lapse imaging, an in vitro Matrigel model of PNI that assesses interactions between nerves and cancer cells, an in vivo murine sciatic nerve model of PNI, and studies of surgically resected human cancers. The combination of these methods will position us to achieve three Aims: 1. Identify putative mechanisms of cancer cell chemotaxis towards GDNF. We will focus on: (a) RET-GFR1 receptor expression, and the RET G681S polymorphism, (b) PI3K/PTEN activity and the p1101 H1047R and E545K mutations, (c) Rho family (Cdc42, Rac1, RhoA) and Ras GTPase activity. 2. Define the role of chemotactic mechanisms in models of cancer cell PNI. Using in vitro and in vivo models, we will correlate modulation of the factors in Aim 1 with changes in PNI phenotype. 3. Validate identified chemotactic mechanisms in human cancers. We will use surgical specimens for molecular, pathologic, and clinical correlations to validate the identified factors from Aims 1 & 2. This proposal will allow us to (1) define mechanisms of cancer cell chemotaxis in perineural invasion, (2) predict PNI behavior based on an assessment of cancer cell characteristics, (3) identify novel targets for therapy by chemotactic inhibition, and (4) introduce a novel therapeutic approach in targeting a cancer's interactions with its microenvironment to modify its behavior, rather than targeting its viability.