PROJECT SUMMARY / ABSTRACT Choosing the more effective therapy is imperative for borderline-resectable pancreatic cancer (BRPC). BRPC is a subcategory of pancreatic adenocarcinoma (PDAC) (about 20% of the entire PDAC cases). BRPC contacts peripancreatic arteries and/or veins but has the potential to be successfully resected after downstaging with effective neoadjuvant therapy. Currently, there are two first-line therapeutic options for BRPC patients, nab- paclitaxel with gemcitabine and FOLFIRINOX. Overall response rates of these regimens are comparable (20- 30%), and there are no robust data favoring one over the other. Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) has the potential as a non-invasive tool for early evaluation of PDAC response to chemotherapy. PDAC is typically hypo-perfused due to significant tumor sclerosis creating elevated interstitial pressure and consequently compressing tumor-feeding vessels. However, an effective therapy induces acute necrosis, reducing interstitial pressure and increasing perfusion. As DCE-MRI can measure tissue perfusion by monitoring the dynamic change of MRI contrast agents introduced intravenously, it can detect perfusion increase in responding PDAC before the morphological change of tumors. However, variability in quantitative DCE-MRI measurement remains a concern. We developed a perfusion phantom named P4 (Point-of-care Portable Perfusion Phantom) to reduce the variability in quantitative DCE-MRI measurement. The reproducibility of quantitative DCE-MRI measurement (e.g., volume transfer constant: Ktrans) of various abdominal tissues across three MRI scanners significantly increased after P4-based error correction (Intraclass correlation coefficient: 0.39 vs. 0.98). We demonstrated that quantitative DCE-MRI could be used to identify the early therapeutic response of PDAC after P4-based error correction. In our pilot study, DCE-MRI was applied for 20 PDAC patients before and 6-8 weeks after therapy initiation. The Ktrans of pancreatic tumors favorably responding to chemotherapy increased 84±26% (n=11) after P4-based error correction, while that of non-responding tumors did not (-7±11%) (n=9) (p<0.0001). We achieved 100% accuracy in differentiating between responding and non-responding tumors. We hypothesize that DCE-MRI-guided neoadjuvant chemotherapy will improve the negative margin (R0) resection rate for BRPC patients when the P4 is used for correcting errors in the imaging data. We propose to switch the first-line therapeutic regimen to the second one if the Ktrans in the tumor increases more than the threshold. We will compare the R0 resection rate of the group that received DCE-MRI-guided neoadjuvant chemotherapy with that of the historical control group that received standard-of-care treatment (primary endpoint). Also, we will determine whether quantitative DCE-MRI can be a reliable tool for assessing pancreatic tumor microenvironment after P4-based error correction using digi...