Abstract P4-02-05: Indocyanine green (ICG) fluorescence mapping for sentinel lymph node (SLN) localization in early breast cancer

Abstract

Abstract Background:Dual localization methods with blue dye and radioisotope are commonly employed for SLN identification but allergic reactions together with staining of skin and surgical tissues disadvantage blue dye. A previous feasibility study using blue dye, radioisotope and ICG confirmed high sensitivity of ICG fluorescence for SLN detection with three-quarters of nodes both radioactive and fluorescent.1 This follow-on study has evaluated a combination of ICG with radioisotope for SLN identification. Methods: In a prospective observational study, 79 patients with unilateral clinically node negative tumors scheduled to undergo routine SLN biopsy for core-biopsy proven invasive (n= 78) or non-invasive (n = 1) breast cancer were identified at multidisciplinary meetings [43 screen-detected; 36 symptomatic]. All patients received dual localization with radiocolloid (Technetium nanocolloid, 20MBq) and ICG (0.5%) and had pre-operative axillary ultrasound. The number of nodes was recorded numerically and whether radioactive, fluorescent or both. Lymphatic and nodal tissue were visualized with a Photodynamic Eye camera and sensitivity of individual tracers alone and in combination calculated. Institutional approval was granted for use of ICG instead of blue dye. Statistical analysis was performed using Stata (version 15.1) with student t-test for normal distribution and Mann-Whitney test for other variables. The main objective was to assess non-inferiority of fluorescent (ICG) compared with standard radioisotopic localization. This was based on a confidence interval for the difference in percentage detection rate with a non-inferiority margin of 5%. Results: A total of 162 nodes were retrieved from 79 patients with an average nodal count of 2.04 (range 1 - 4) and an overall identification rate of 98.7% (78/79). Amongst these excised nodes, 154 displayed uptake of tracer and were either fluorescent, radioactive or both (8 nodes were removed incidentally or were palpably suspicious and tracer negative). More than 90% (151/154) of nodes were fluorescent and 73.4% (113/154) were radioactive with at least 10% activity of the hottest node. These results yielded an overall concordance rate of 71.4%. Nodal detection rates for ICG alone or combined with radioisotope were 98.1% (151/154) and 73.4% (113/154) respectively whilst procedural detection rates were 97.5% (77/79) for radioisotope and 98.7% (78/79) for ICG. Metastases were present in 13 nodes (all fluorescent and hot) with 13 patients having a single positive node containing macrometastases (n= 5), micrometastases (n = 6) or isolated tumor cells (n = 2). The node positivity rate was 14.1% for macro- or micrometastases and 2.5% for isolated tumor cells. ICG was non-inferior to radioisotope for both procedural and nodal detection rates with the lower confidence interval not crossing within the pre-defined limit. No serious adverse reactions were recorded. Conclusion: ICG fluorescence imaging permits real-time visualization of lymphatics and provides an additional dimension to SLN biopsy that is safe and effective. These results confirm high sensitivity for fluorescence localization with comparable performance to the gold standard using radioisotope. ICG can reliably be employed as a sole tracer that avoids potential drawbacks of standard tracer agents including availability and costs of radioisotope. 1. Wishart GC, Jones LC, Loh S-W, Benson JR. A feasibility study (ICG-10) of indocyanine green (ICG) fluorescence mapping for sentinel lymph node detection in early breast cancer. Eur J Surg Oncol 2012; Jun 13 Citation Format: Dorin Dumitru, Sujit Gnanakumar, Elena Provenzano, John Benson. Indocyanine green (ICG) fluorescence mapping for sentinel lymph node (SLN) localization in early breast cancer [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr P4-02-05.