• Neuroedocrine tumor
• Neuropilin
• pancreas
• PlGF
• prognosis
• Robo
• Slit

Background: pancreatic neuroendocrine tumors (pNETs) prognosis is difficult to predict. In
other tumor models Placental growth factor (PlGF)-Neuropilin (NP) and Slit-Robo signaling have been involved in angiogenesis and tumor spread, but their function in pNETs and their potential prognostic role need to be clarified.
Methods: a clinical evaluation was initially performed into two series of pNET patients, correlating PlGF, Neuropilin 2 (NP2), Slit2 and Robo1 tumoral expression to overall survival, progression free survival (PFS) and main clinical features.
An Italian cohort of 55 surgically treated pNETs was retrospectively investigated, evaluating PlGF and NP2 expression by immunohistochemistry (IHC) technique. Moreover,PlGF-NP2 IHC expression was evaluated in a smaller series of 12 advanced pNETs treated with Everolimus, and its expression was retrospectively correlated to clinical response to that target therapy.
In a German series of 45 pNET, Slit2 expression was assessed by qPCR, while Robo1 expression was evaluated by qPCR and IHC, and compared with the one observed in 27 normal pancreatic tissue samples.On the basis of the clinical results, Slit2-Robo1 pathway was further investigated in vitro.Migration, cell proliferation and colony formation (CF) were evaluated in BON and QGP cell lines with and without Slit2 and Robo1 expression.
Results: NP2 and PlGF expression was not significantly associated with main clinical and
biological features, survival or PFS, in the evaluated pNET series.
On the contrary, a reduced Robo1 expression was associated with a shorter time to tumor
progression (TTP) (p=0.025). Slit2mRNA, although not significantly correlated with survival or TTP, appeared significantly reduced in pNETs, as compared to normal pancreas(p=0.0019). From the in vitro studies, Slit2 expression in BON cells revealed to inhibit proliferation,migration and CF, while Robo1 knockdown or sequestration of Slit2 in QGP,stimulated migration and CF.
Conclusion: PlGF-NP expression seemed not helpful to predict pNETs prognosis.On the opposite, Slit2-Robo1 showed a tumor-suppressor function in vitro and Robo1 loss was related to a worse clinical prognosis, suggesting its potential role as prognostic marker in pNETs.