Multisequence MRI With Functional Imaging May Improve Pseudoprogression vs. Viable Tumor Determination Following High Dose Adaptive MRgRT in Patients With Pancreatic Adenocarcinoma

Abstract

<h3>Purpose/Objective(s)</h3> For patients with Borderline Resectable (BR) and Locally Advanced (LA) pancreatic adenocarcinoma, multi-fraction ablative stereotactic body radiotherapy (SBRT) by MRgRT provides a highly effective approach for locoregional therapy. Traditionally, the first post treatment scan is performed with multi-phase CT, but a multiparametric and multisequence MRI may offer a better approach to evaluate post-treatment response using a reproducible exam with functional imaging and quantifiable biomarkers. The objective of this study is to develop MR biomarkers to more accurately and reproducibly evaluate treatment response and to differentiate between post-treatment inflammatory response resulting in pseudoprogression from background post-treatment fibrosis and residual disease. <h3>Materials/Methods</h3> With institutional review board approval, patients with BR and LA pancreatic adenocarcinoma who were imaged with MRI 3-5 months following MRgRT with pre- and post-SBRT pancreas protocol CT scans were included. Primary endpoints were to assess for treatment response, to identify residual foci of disease, and to identify pseudoprogression caused by inflammation, fibroinflammatory change, post-treatment fibrosis, and necrotic change. A specific diagnostic MRI protocol was developed at our institution to evaluate for post treatment changes. <h3>Results</h3> Ten patients satisfied the above inclusion criteria. Employing a morphologic approach with traditional RECIST criteria, all patients demonstrated an increase in the size of the primary tumor between pre-treatment and post treatment CT scans (an average increase of 24 % +/- 13.4%). On MRI these post-treatments beds demonstrated functional imaging signs of response with an average apparent diffusion coefficient (ADC) above a threshold level of 1500 µm²/sec (average of 1993.9 µm²/sec +/- 437), consistent for necrotic change or cystic debris. Additionally, the normalized tumor contrast ratio measured below 0.5 (average of 0.194 +/- 0.13), consistent for no appreciable internal enhancement with kinetic profiles showing no rapid wash in or contrast persistence. Functional T2 map values were greater than 150 ms (average of 217 ms +/- 81.2), highlighting residual inflammation, while functional T1 map values were obscured by this same inflammatory signal (range, 1036 -3966 ms; average of 1930 +/- 1143). <h3>Conclusion</h3> Specialized post-SBRT multisequence MRI protocols with functional imaging detail pseudoprogression that is frequently characterized by CT as stable or worsening disease. These findings have important implications for assessment of post SBRT treatment response and further therapy options. Further data is needed to clarify the kinetics of response to determine the most appropriate post treatment imaging acquisition time point. Future validation with prospective post SBRT MRI imaging with specific sequencing protocols to assess response is warranted.