Sentinel node mapping in endometrial cancer: A Brazilian cancer center experience

Abstract

Objectives: The role of sentinel lymph node (SLN) mapping in endometrial cancer is still in debate. Our aim was to determine the detection rate, sensitivity, and negative predictive value of SLN mapping in low- and high-risk (grade 3, serous, clear cell, deep invasion, or angiolymphatic invasion) endometrial cancer. Methods: We analyzed a series of 91 patients treated at AC Camargo Cancer Center from January 2013 to August 2015 who underwent SLN mapping with cervical injection of blue dye. Forty-nine (53.8%) patients had low-risk tumors and no further lymph node dissection (LND) or only pelvic LND. Forty-two (46.2%) patients had high-risk tumors and received pelvic ± para-aortic LND. Results: Median age was 59.7 years (range, 37-84 years). Median body mass index (BMI) was 28.1 (range, 18-54). Twenty-one patients (23.1%) had open surgeries and 70 underwent (76.9%) minimally invasive surgeries. Thirty-two patients (36.8%) had pelvic LND, and 27 (28.2%) pelvic + para-aortic LND. The rate of SLN detection was 79.1% overall, and bilateral in 50.5% of cases. The median SLN detected was 2 per patient (range, 1-8). Sixteen patients (18.2%) had histologic grade 3, 17 (18.7%) had angiolymphatic invasion, 8 (8.8%) had serous/clear cell histology, and 22 (24.2%) had deep myometrial invasion. Notably, endometrial cancer was diagnosed after subtotal hysterectomy in 3 cases, and bilateral SLN was detected in all. BMI, age, open surgery, blue dye volume (2 vs 4 mL), and first 30 cases did not influence the detection rate. Four patients (4.4%) had technical difficulty with cervical injection, and in this situation, SLN was detected in only 1 (25%) case. No positive node was found in low-risk cases. In high-risk cases, 4 patients had only SLN with no further LND for medical reasons. In the high-risk group, at least 1 SLN was found in 34 (81%) patients and was positive in 8 (23.5%) cases. Two (5.8%) cases had isolated tumor cells, 1 (2.9%) had micrometastasis, and 5 had (14.7%) macrometastasis. Notably, 3 (37.5%) of 8 SLN metastases were found only on immunohistochemistry. Only 1 patient had false-negative SLNs, with ipsilateral pelvic and para-aortic non–sentinel positive nodes. The sensitivity was 88.9%, negative predictive value was 97.4%, and false-negative rate was 11.1%. Two (25%) patients with positive SLN had also other pelvic and para-aortic positive non-SLN. Conclusions: Our data suggest that SLN mapping is a safe and accurate technique for high-risk tumors, and increases metastatic detection rate by 8.7% in high-risk tumors.