86 (PB-086) Poster - Magnetic delayed sentinel lymph node dissection in primary systemic therapy. Implications for enhanced axillary mapping (Update)

Abstract

Background: Superparamagnetic Iron Nanoparticles (SPIO) have been used for Sentinel Lymph Node Detection in breast cancer with equivalent efficacy to Radioisotope (RI) ± blue dye while facilitating logistics and leading to successful SLND even when injected up to eight weeks before surgery. However, neither the role of SPIO for SLND after primary systemic therapy (PST) nor the maximum timeframe from SPIO administration to successful SLND have been adequately defined. Methods: Patients originally cN0/1 planned for PST were included. All with cN0 axilla underwent SLND while patients with cN1 who converted to ycN0 received targeted axillary dissection (TAD) with a paramagnetic clip. All patients received SPIO either before PST or before surgery and RI was injected on the day of surgery. Results: In total, 114 patients were included in the analysis (median age 56 yrs; iqr 45, 68, median BMI 25.1 kg/m2; iqr 22.9, 28.8), of which 91 (79.8%) received chemotherapy ± targeted therapies and 23 (20.2%) endocrine therapy. SLND was performed in 81 (71.1%) and TAD in 33 (28.9%) of patients. SPIO was injected within a week to surgery in 74 (64.9%) patients and longer than a week before in 40 (35.1%), in a median of 2 days (range 0–248) for the entire cohort. At least one SLN was detected in 97.4% with SPIO and 92.1% with RI (p = 0.149) while the combination was successful in 100%. At least one SLN was concordant for SPIO and RI in 83.3% of patients. The median (iqr) lymph node yield was 3 (2,4) for SPIO, 2 (2,3) for RI and 3 (2,4) for the combination (p < 0.001). Time from SPIO injection to surgery affected detection (Spearman’s rho: 0.194, p = 0.039) but not number of SLNs (rho: −0.062, p = 0.511) or concordance per patient (rho: −0.057, p = 0.544). The addition of SPIO to RI significantly increased overall detection rate (difference 7.9%, p = 0.008), but the addition of RI to SPIO did not significantly improve overall detection (difference 2.6%, p = 0.248). In patients undergoing TAD (n = 33), detection was 100% for SPIO and 90.9% for RI (p = 0.248). The index node was magnetic in 89.3% and radioactive in 64.3% (p = 0.035), an outcome not affected by any factors. A median (iqr) of 1 (1,2) axillary metastases were found in 25 patients (21.9%). SPIO detection was 100% and RI detection was 70.8% (p = 0.023). SPIO detected more metastatic SLNs than RI (median[iqr] 1 (1,1) vs 1 (0,1); p = 0.005). In completion ALND, additional metastatic nodes were found in one patient (4%). Conclusions: In this well-defined single-arm cohort study, SPIO performed comparably to RI, but detected more SLNs and had higher detection of metastatic SLNs. Injection before PST is not only feasible, but does not seem to affect concordance with RI. These findings support the use of SPIO in PST patients and motivate more dedicated research in the concept of delayed SLND in this setting. No conflict of interest.