Sentinel lymph node assessment in oesophageal cancer: creation of a multimodality tracer

Abstract

Introduction Early oesophageal cancer can now be treated endoscopically; however, more accurate lymph node assessment is required. The sentinel lymph node (SLN) is defined as the first lymph node to which the primary tumour drains. This node should reflect the status (i.e. benign or malignant) of the entire lymph node basin. Our aims were to identify current difficulties that limit the acceptance of the SLN concept in oesophageal cancer, as well as to create a multimodal (magnetic resonance imaging [MRI] and gamma) tracer, to enable more detailed perioperative SLN assessment. Methods A literature review was undertaken, targeting studies assessing oesophageal cancer and SLN. MEDLINE, the Cochrane Database of Systematic Review, as well as PUBMED, were consulted. The keywords and medical subject headings (MeSH) used were ‘(o)esophageal cancer’, ‘(o)esophageal adenocarcinoma’, ‘(o)esophageal squamous cell carcinoma’ and ‘SLN’ used in combination with AND or OR. Only studies in English were considered. ⁹⁹ᵐTc-­‐labelled magnetic nanoparticle formulations were then investigated for further use as a multimodality SLN contrast agent. Radiolabelling of the dextran-­‐coated magnetic nanoparticles was undertaken, and instant thin layer chromatography developed to assess efficiency. Bio- distribution was determined after intravenous and subdermal injection in the tails of Sprague-­‐Dawley rats. Lymphoscintigraphy was performed on the subdermally injected rats to assess lymphatic mapping. Following this, the multimodality SLN tracer was injected into the oesophagus of four swine. MRI images were then acquired, and then sentinel nodes were further assessed intraoperatively with a gamma probe. Bio-­‐distribution was then assessed, comparing ⁹⁹ᵐTc-­‐antimony trisulphide colloid (ATC) and ⁹⁹ᵐTc-­‐superparamagnetic iron oxide nanoparticles (SPIONs). Lymph nodes and reticuloendothelial organs were then harvested and counted to determine the percentage of injected dose (%ID). Results The literature review demonstrated at least four issues limiting the use of the SLN concept in oesophageal cancer. These included: timing of tracer – different radiocolloids overseas, with different rates of flow; the chest cavity – difficulty of access post-­‐injection; blue dye – not useful, with proven false-­‐negative rate; and shine-­‐through effect. The multimodality tracer formulation was optimised, with labelling efficiency confirmed at >99%. In the Sprague-­‐Dawley rats, the tracer was injected subdermally, with lymphoscintigraphy demonstrating excellent delineation of the sentinel nodes. The reticuloendothelial organs were then removed and activity measured, with the liver showing 3.2% highest mean activity, followed by the kidneys with 2%, the spleen at 0.3% and the lungs at 0.1%. The gamma probe detected all nodes identified on MRI imaging of the four swine injected with ⁹⁹ᵐTc-­‐SPIONs. Radiolabelling efficiency was >98%. ⁹⁹ᵐTc-­‐ATC and ⁹⁹ᵐTc-­‐ SPIONs were taken up by swine liver and lungs with similar percentage ID values, and there was 7% ID of ⁹⁹ᵐTc-­‐SPIONs by the kidneys, compared to <1% ID of ⁹⁹ᵐTc-­‐ATC in the same organs. Conclusion There are numerous issues that require evaluation and agreement in the international community to ensure greater use of the SLN concept in oesophageal cancer. A multimodality, ⁹⁹ᵐTc-­‐SPIONs, tracer was validated using both MRI imaging and a gamma probe. This tracer requires further assessment, as it could allow pre-­‐ operative assessment of sentinel lymph nodes in early oesophageal cancer.