Sentinel lymph node biopsy in early stage endometrial cancer: a Turkish gynecologic oncology group study (TRSGO-SLN-001)

Abstract

Objective The aim of this multicenter study was to evaluate the feasibility of sentinel lymph node (SLN) mapping in clinically uterine confined endometrial cancer. Methods Patients who underwent primary surgery for endometrial cancer with an SLN algorithm were reviewed. Indocyanine green or blue dye was used as a tracer. SLNs and/or suspicious lymph nodes were resected. Side specific lymphadenectomy was performed when mapping was unsuccessful. SLNs were ultrastaged on final pathology. Results 357 eligible patients were analyzed. Median age was 59 years. Median number of resected SLNs was 2 (range 1-12) per patient. Minimal invasive and open surgeries were performed in 264 (73.9\%) and 93 (26.1\%) patients, respectively. Indocyanine green was used in 231 (64.7\%) and blue dye in 126 (35.3\%) patients. The dyes were injected into the cervix in 355 (99.4\%) patients. The overall and bilateral SLN detection rates were 91.9\% and 71.4\%, respectively. The mapping rates using indocyanine green or blue dye were comparable (P=0.526). There were 43 (12\%) patients with lymphatic metastasis. The SLN algorithm was not able to detect 3 of 43 patients who had isolated paraaortic metastasis. After SLN biopsy, complete pelvic lymphadenectomy was performed in 286 (80.1\%) patients. Sensitivity and negative predictive value were both 100\% for the detection of pelvic lymph node metastases. In addition, 117 (32.8\%) patients underwent completion paraaortic lymphadenectomy after SLN biopsy. In these patients, sensitivity for detecting metastases to pelvic and/or paraaortic lymph nodes was 90.3\% with a negative predictive value of 96.6\%. The risk of non-SLN involvement in patients with macrometastatic SLNs, micrometastatic SLNs, and isolated tumor cells in SLNs were 61.2\%, 14.3\% and 0\%, respectively. Conclusions SLN biopsy had good accuracy in detecting lymphatic metastasis. However, one-third of cases with metastatic SLNs also had non-SLN involvement and this risk increased to two-thirds of cases with macrometastatic SLNs. The effect of leaving these nodes in situ on survival should be evaluated in further studies.