PROJECT SUMMARY/ABSTRACT: Neuroendocrine tumors (NETs) are rare and slow growing malignancies with an increasing incidence and prevalence1. The relevance of current standard imaging modalities and response assessment criteria (RECIST 1.1) is questioned for evaluation of NETs due to inter-observer variability and the inability to evaluate early response to therapy (changes in vascularity, textures, density)4. Ours is a unique pilot project where we utilize and repurpose an untapped resource i.e., image bank created from the cooperative group study E2211 that randomized patients with pancreatic NETs to capecitabine +/- temozolamide therapy. We will review all of the images available in the image bank and perform secondary analysis to quantify variability in both: imaging technique/radiation dose (discrepancies in image quality) and tumor measurements using RECIST 1.1 (response criteria). These findings are vital to disease evaluation in NETs. Alongside, we also hypothesize that there are alternate criteria (RECIST lite, Choi, changes in texture, intensity and other quantitative imaging markers) that could evaluate response to therapy on CT imaging better and earlier than the current standard (RECIST 1.1) as previously shown in gastrointestinal stromal tumors (Choi criteria)5,6. The feasibility of image banking and its utility in answering clinically relevant questions is also being indirectly evaluated through our study. Assessing the quality of imaging studies and quantifying the inter-reader variability will highlight the importance of standardizing disease assessment tools and justify the need for exploring alternate tools. Identification of alternate radiological criteria and quantitative imaging biomarkers that are early predictors of response will help us identify patients who are more likely to respond to any specific therapy.