Preclinical evaluation of a dual magnetic and fluorescence-guided approach for sentinel lymph node biopsy in the stomach.

Abstract

337 Background: Gastrectomy with extended (D2) lymphadenectomy is considered as a standard of care for gastric cancer to provide best possible outcomes and pathologic staging. However, D2 gastrectomy is technically demanding operation and reported to be associated with increased complications, including bleeding, pancreatitis, abscess, chylous ascites, and mortality. Application of sentinel lymph node mapping in gastric cancer has the potential to reduce patient morbidity, while facilitating detailed pathology processing required to accurately identify nodal metastases. However, sentinel lymph node biopsy techniques are not established for gastrectomy because of a lack of practical tracers; an effective and convenient tracer that exclusively accumulates in sentinel lymph nodes is critically needed. Methods: A lymphotropic tracer solution consisting of mannose-labelled magnetic nanoparticles (SPIONs) and a fluorescent dye (indocyanine green, ICG) was injected laparoscopically into the stomach submucosa of 6 healthy female domesticated pigs while under general anaesthesia. Intraoperative laparoscopic fluorescence imaging was used to highlight draining lymphatic pathways containing ICG, while a combination of T2-weighted MRI and magnetometer probe detection were used to identify nodes containing SPIONs. Results: Mixing the ICG dye and SPIONs ensured concurrence of fluorescent and magnetic signals in the sentinel nodes. Average tracer accumulation time before dissection was 5 hours. On average 4.5 fluorescent nodes (range 2–7) were removed, of which 2 (average) were also magnetic (range 1–4). Magnetic nodes could be identified using both preoperative MRI and intraoperative magnetometer probe. The mannose label of the SPIONs produced a strong binding affinity to macrophages, and in combination with particle size (80 nm) results in enhanced retention in the sentinel nodes. On average, after 5 hours accumulation time, the sentinel nodes contained 4.1% of the SPION injected dose and 88% of the magnetometer counts. Conclusions: Through the co-injection of fluorescent and mannose-labelled magnetic tracers, lymphatic drainage pathways can be determined preoperatively with MRI, visualised with intraoperative fluorescence cameras, and SLNs confirmed with magnetometer. Although ICG flows through multiple echelons, in doing so it creates a lymphatic roadmap to guide surgeons. The addition of magnetic particles provides sentinel node discrimination, so in combination, these two tracers could greatly enhance the capability to detect SLNs in gastric cancer.