PROJECT SUMMARY / ABSTRACT The overall goal of this study is to validate a UAB-invented perfusion phantom named P4 (Point-of-care Portable Perfusion Phantom) for accurate DCE-MRI measurement of pancreatic tumor response to neoadjuvant therapy. DCE-MRI has been utilized to noninvasively assess early therapeutic response in solid tumors by detecting changes in tissue perfusion prior to anatomic tumor shrinkage. However, measurement variability across different scanners remains a major concern in multi-site trials of DCE-MRI. To address this concern, we invented the P4 that is small enough to be imaged concurrently in the bore of a standard MRI scanner with a patient, thus can serve as an internal reference to detect and correct the scanner-dependent variation in quantitating DCE-MRI parameters. We have confirmed that the use of this device decreased variability in DCE-MRI measurement to about 4%, although it was about 20% before correction. We now hypothesize that the P4 can help identify early therapeutic response of pancreatic tumors in the neoadjuvant setting. Pancreatic tumors are typically hypo-perfused, but we have recently demonstrated that the tumor perfusion can be significantly increased after an effective chemotherapy. Choosing the more effective therapy early is particularly important for patients with borderline-resectable pancreatic cancer, as an effective therapy can downstage a tumor to allow curable surgery. Three specific aims are proposed as follows. Aim 1 (UG3- 1): Develop a disposable P4 that is concurrently imaged with a patient for improved accuracy and reproducibility in quantitative DCE-MRI measurement. For routine and widespread clinical use, the phantom should be ready-to-use, inexpensive and disposable. Injection molding methodology will be employed to manufacture the disposable P4. Aim 2 (UG3-2): Develop a software package analyzing abdominal DCE-MRI images obtained with the P4. Image processing will be conducted sequentially in eight steps. To date, we have developed eight prototype software program modules (one for each step). In this study, we aim to combine all eight modules for seamless data flow, while making each module automated/semi-automated not only to reduce user bias but to improve speed of operation. The software front panel will be updated for more efficient operation using feedback from multiple users. Aim 3 (UH3-1): Evaluate the changes of DCE-MRI parameters as surrogate imaging biomarkers for pancreatic cancer therapy response in a multi-site setting after error correction using the P4. A total of 50 patients with borderline resectable pancreatic cancer entering neoadjuvant therapy will be recruited in two research institutes (n=25 per institute). The changes of DCE-MRI parameters in tumors for 4 weeks after therapy initiation will be correlated with the therapeutic response assessed by margin-negative resection rate after completion of neoadjuvant therapy, and evaluated as surrogate biomarkers after P4-based error correction.

PROJECT NARRATIVE The purpose of this project is to validate a UAB-invented perfusion phantom as a clinical tool to measure pancreatic tumor response to therapy early and accurately in dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). During UG3 phase, a disposable perfusion phantom and a data analysis software package will be developed. During UH3 phase, the clinical utility of those will be validated to improve the accuracy of DCE-MRI measurement for early assessment of pancreatic tumor response to therapy.