Targeted near-infrared imaging utilizing a cathepsin-activated fluorophore for the intraoperative detection of canine insulinoma.

Abstract

OBJECTIVES: The aim of this study was to evaluate the use of cathepsin-activated intraoperative near-infrared (NIR) imaging to detect insulinomas in dogs, a spontaneous large animal model for human disease. MATERIALS AND METHODS: A prospective, pilot clinical trial was performed on dogs with naturally occurring insulinomas undergoing exploratory laparotomy. Each dog underwent routine preoperative diagnostic assessment, and a cathepsin-activated fluorophore (VGT-309) was administered intravenously 1-2 days preoperatively. All intraoperative findings with visible light and NIR imaging were recorded and mean NIR fluorescence intensity of tumors and grossly normal pancreas were quantified. Excision of any identified primary tumor and suspected metastatic lesions was performed. All excised tissues underwent histologic evaluation and immunohistochemistry (IHC) for cathepsin B expression. Descriptive statistics were calculated, and differential fluorescence intensity and cathepsin B expression between the pancreatic mass and adjacent grossly normal pancreatic tissue were assessed for statistical significance via paired t tests with p&#x2009;<&#x2009;0.05 used for significance. RESULTS: Six dogs were enrolled. No adverse events occurred secondary to administration of the imaging agent. In situ, insulinomas had significantly greater mean fluorescence intensities than the surrounding pancreas, and the median tumor to background ratio was 1.906 (range 1.286-2.556). One dog had an occult pancreatic mass that was identified intraoperatively with NIR guidance. Background fluorescence of liver and lymph nodes was observed in all cases, and one dog was diagnosed with nodal and hepatic metastasis. Histologic tumor margins correlated with margins of NIR fluorescence. Cathepsin B expression was determined to be significantly greater in the pancreatic tumor compared to adjacent non-neoplastic pancreas via IHC, and there was no overlap in the range of median IHC-positive proportion values for these tissues. However, there was overlap in the range of IHC-positive proportion values for neoplastic pancreatic samples and lymph node and liver tissues. CLINICAL SIGNIFICANCE: The findings of this pilot study support further investigation of cathepsin-activated NIR imaging to enhance intraoperative canine insulinoma localization and margin evaluation. Future studies are needed to further characterize and optimize the utility of targeted NIR imaging, particularly to identify metastatic lesions, for canine insulinoma, which may serve as an effective translational model for humans with pancreatic neuroendocrine tumors.