The primary reason for a poor prognosis for patients with pancreatic cancer is because 80-90% of patients have unresectable disease at the time of diagnosis. This research proposal offers a strategy that may enable the detection of pancreatic cancer by MRI with at least two to three orders of magnitude smaller than those currently detected by employing target specific hyperpolarized silicon nanoparticles. This strategy when successfully introduced in the clinic will improve the overall detectability of pancreatic tumors a early stages of their development and could potentially save lives by successful surgical interventions. Among pancreatic tumor biomarkers produced in pancreatic tissue, the hepatocarcinoma-intestine-pancreas/pancreatitis- associated protein (HIP/PAP) was found to be over-expressed more than 130-fold in pancreatic acinar cells in pancreatic cancer, as compared to normal pancreas. In contrast, only a 9-fold increased expression of HIP/PAP protein was observed in acinar cells in chronic pancreatitis. This offers a unique opportunity to image the tumor and tumor stromal interaction in pancreatic cancer. The objective of this exploratory project is to develop a highly sensitive HIP/PAP targeted molecular imaging agent based on a novel silicon hyperpolarized Magnetic Resonance Imaging (MRI) technique recently developed in my laboratory. Previous biochemical studies demonstrated that HIP/PAP protein is a "lactose binding protein" which has a very high affinity to D- lactose. We intend to employ lactose-functionalized hyperpolarized Silicon nanoparticles (SiNPs) to target HIP/PAP protein in the acinar cells surrounding the pancreatic tumor. Hyperpolarization leads to over 10,000 fold signal enhancement compared to conventional MRI. We hypothesize that the specificity of lactose- functionalized SiNPs binding to HIP/PAP surrounding the pancreatic cancer coupled with the sensitivity gain due to hyperpolarization will lead to an "effective lesion size amplification", thereby, enabling the visualization of pancreatic carcinoma in in vivo MRI with at least two to three orders of magnitude smaller than those currently detected. Towards this goal, we propose the following three specific aims: I. Optimize the levels of hyperpolarization achievable by radical-free 29Si Dynamic Nuclear Polarization on lactose-functionalized silicon nanoparticles (SiNPs) as a function of SiNP size and in vivo targeting efficacy. II. Demonstrate "effective lesion size amplification" with real-time hyperpolarized MRI in a) one orthotropic pancreatic tumor xenograft mouse model and in b) two cohorts of patient derived tumor engrafted into NOD/SCID mice one of which is na�ve human pancreatic tumor and the other exposed to neoadjuvant therapy. III. Correlate hyperpolarized 29Si MR data acquired in Aims II with immunohistochemistry and optical imaging data of fluorescently-tagged SiNPs in the murine cancer models demonstrating efficacy of this technique on the detection pancreatic carcinoma with at least two to three orders of magnitude smaller than those currently detected.