Multiwavelength fluorescence imaging, combining ICG with a tumor-specific tracer, may overcome these limitations.
This prospective feasibility study therefore evaluated the feasibility of simultaneous multiwavelength fluorescence imaging using SGM-101 in addition to standard ICG-guided CRLM surgery.
Ten patients with high-risk features for R1 resection received ICG (0.5 mg/kg) 7 ± 1 days and SGM-101 (10 mg) 4 ± 1 days before surgery.
The prespecified sensitivity criterion (4) was not met, with an overall SGM-101 sensitivity of 50% on a lesion-level and 70% on a patient-level.
These findings support further evaluation of multiwavelength fluorescence imaging in larger clinical trials.
Complete (R0) resection is essential for curative treatment of colorectal liver metastases (CRLM). Fluorescence-guided surgery with indocyanine green (ICG; emission wavelength ~ 800 nm) is increasingly being adopted as standard care because it enhances lesion detection and improves R0 resection rates. However, unintended R1 resections still occur, while the limited tumor specificity of ICG may lead to false-positive resections. Multiwavelength fluorescence imaging, combining ICG with a tumor-specific tracer, may overcome these limitations. SGM-101, a carcinoembryonic antigen-targeting fluorescent antibody (emission wavelength ~ 680 nm), enables tumor-specific visualization of colorectal cancer cells. This prospective feasibility study therefore evaluated the feasibility of simultaneous multiwavelength fluorescence imaging using SGM-101 in addition to standard ICG-guided CRLM surgery. Ten patients with high-risk features for R1 resection received ICG (0.5 mg/kg) 7 ± 1 days and SGM-101 (10 mg) 4 ± 1 days before surgery. Feasibility was predefined by four criteria: surgical usability and patient experience assessed through surveys, absence of spectral interference, and SGM-101 sensitivity > 80% for CRLM detection, considered necessary for potential intraoperative R1-margin identification. Three of the four predefined feasibility criteria were met. (1) Surgeons reported excellent usability; (2) patients experienced no burden from additional tracer administration, and (3) no spectral interference was observed. The prespecified sensitivity criterion (4) was not met, with an overall SGM-101 sensitivity of 50% on a lesion-level and 70% on a patient-level. However, in an exploratory subgroup analysis, sensitivity increased to 90% in lesions with higher tumor cell density (tumor regression grade 3–5). SGM-101’s specificity for lesion and margin discrimination was 100%. Importantly, all R1 margins were identified intraoperatively by at least one of the two tracers. These findings support further evaluation of multiwavelength fluorescence imaging in larger clinical trials. Refining preoperative patient selection may be key to improving sensitivity in future studies.
