NIR-II fluorescence imaging-guided colorectal cancer surgery targeting CEACAM5 by a nanobody

Background Surgery is the cornerstone of colorectal cancer (CRC) treatment, yet complete removal of the tumour remains a challenge. The second near-infrared window (NIR-II, 1000-1700 nm) fluorescent molecular imaging is a novel technique, which has broad application prospects in tumour surgical navigation. We aimed to evaluate the ability of CEACAM5-targeted probe for CRC recognition and the value of NIR-II imaging-guided CRC resection.Methods We constructed the probe 2D5-IRDye800CW by conjugated anti-CEACAM5 nanobody (2D5) with near-infrared fluorescent dye IRDye800CW. The performance and benefits of 2D5-IRDye800CW at NIR-II were confirmed by imaging experiments in mouse vascular and capillary phantom. Then mouse colorectal cancer subcutaneous tumour model (n = 15), orthotopic model (n = 15), and peritoneal metastasis model (n = 10) were constructed to investigate biodistribution of probe and imaging differences between NIR-I and NIR-II in vivo, and then tumour resection was guided by NIR-II fluorescence. Fresh human colorectal cancer specimens were incubated with 2D5-IRDye800CW to verify its specific targeting ability.Findings 2D5-IRDye800CW had an NIR-II fluorescence signal extending to 1600 nm and bound specifically to CEACAM5 with an affinity of 2.29 nM. In vivo imaging, 2D5-IRDye800CW accumulated rapidly in tumour (15 min) and could specifically identify orthotopic colorectal cancer and peritoneal metastases. All tumours were resected under NIR-II fluorescence guidance, even smaller than 2 mm tumours were detected, and NIR-II had a higher tumour-to-background ratio than NIR-I (2.55 +/- 0.38, 1.94 +/- 0.20, respectively). 2D5-IRDye800CW could precisely identify CEACAM5-positive human colorectal cancer tissue.Interpretation 2D5-IRDye800CW combined with NIR-II fluorescence has translational potential as an aid to improve R0 surgery of colorectal cancer.Fundings This study was supported by Beijing Natural Science Foundation (JQ19027), the National Key Research and Development Program of China (2017YFA0205200), National Natural Science Foundation of China (NSFC) (61971442, 62027901, 81930053, 92059207, 81227901, 82102236), Beijing Natural Science Foundation (L222054), CAS Youth Interdisciplinary Team (JCTD-2021-08), the Strategic Priority Research Program of the Chinese Academy of Sciences (XDA16021200), the Zhuhai High-level Health Personnel Team Project (Zhuhai HLHPTP201703), the Fundamental Research Funds for the Central Universities (JKF-YG-22-B005) and Capital Clinical Characteristic Application Research (Z181100001718178). The authors would like to acknowledge the instrumental and technical support of the multi-modal biomedical imaging experimental platform, Institute of Automation, Chinese Academy of Sciences.Copyright (c) 2023 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).