Image-Guided Surgery in Patients with Pancreatic Cancer: First Results of a Clinical Trial Using SGM-101, a Novel Carcinoembryonic Antigen-Targeting, Near-Infrared Fluorescent Agent

Abstract

Background: Near-infrared (NIR) fluorescence is a promising novel imaging technique that can aid in intraoperative demarcation of pancreatic cancer (PDAC) and thus increase radical resection rates. This study investigated SGM-101, a novel, fluorescent-labeled anti-carcinoembryonic antigen (CEA) antibody. The phase 1 study aimed to assess the tolerability and feasibility of intraoperative fluorescence tumor imaging using SGM-101 in patients undergoing a surgical exploration for PDAC.

Methods: At least 48 h before undergoing surgery for PDAC, 12 patients were injected intravenously with 5, 7.5, or 10 mg of SGM-101. Tolerability assessments were performed at regular intervals after dosing. The surgical field was imaged using the Quest NIR imaging system. Concordance between fluorescence and tumor presence on histopathology was studied.

Results: In this study, SGM-101 specifically accumulated in CEA-expressing primary tumors and peritoneal and liver metastases, allowing real-time intraoperative fluorescence imaging. The mean tumor-to-background ratio (TBR) was 1.6 for primary tumors and 1.7 for metastatic lesions. One false-positive lesion was detected (CEA-expressing intraductal papillary mucinous neoplasm). False-negativity was seen twice as a consequence of overlying blood or tissue that blocked the fluorescent signal.

Conclusion: The use of a fluorescent-labeled anti-CEA antibody was safe and feasible for the intraoperative detection of both primary PDAC and metastases. These results warrant further research to determine the impact of this technique on clinical decision making and overall survival.

Fig. 1

Tumor and background fluorescence signal (in arbitrary units [AU]) and mean tumor-to-background ratio (TBR) per dose group. The influence of SGM-101 dose and timing on the fluorescence signal and TBR seemed limited in this study

Fig. 2

Fluorescence detection of a primary pancreatic tumor. a Color (left column), fluorescence (middle column), and merged (right column) images from intraoperative imaging of a pancreatic tumor using the Quest imaging system. b Color (left column), fluorescence (middle left column), and merged (middle right column) images of ex vivo imaging of a slice from the same pancreatic tumor using the Quest imaging system and to the Pearl imager (right column). c Histopathologic evaluation and fluorescence signal in a primary pancreatic tumor. Fluorescence microscopy (left column) showing accumulation of SGM-101 in tumor cells. The fluorescence pattern is consistent with carcinoembryonic antigen (CEA) expression measured using immunohistochemistry (IHC, middle column), which corresponds to the site containing tumor cells visible on hematoxylin and eosin (H&E) staining (right column). Note: The acuity of the images is suboptimal compared with the intraoperative setting because these tagged image file format (TIFF) images (8 bits) were subtracted from the intraoperative videos

Fig. 3

Fluorescence detection of a peritoneal and liver metastasis of a pancreatic tumor. a Color (left column), fluorescence (middle left column), and merged (middle right column) images of intraoperative imaging showing a peritoneal metastasis of a pancreatic tumor and images of ex vivo imaging showing a slice from the same metastasis (right column) using the Quest imaging system. b Color (left column), fluorescence (middle left column), and merged (middle right column) images of intraoperative imaging showing a liver metastasis of a pancreatic tumor and images of ex vivo imaging showing a slice from the same metastasis (right column) using the Quest imaging system. Note: The acuity of the images is suboptimal compared with the intraoperative setting because these tagged image file format (TIFF) images (8 bits) were subtracted from the intraoperative videos