Endoscopic Ultrasound Predicts Risk of Occult Intra-Abdominal Metastases in Localized Gastric Cancer: A Validation Study.

Abstract

In gastric cancer (GC) patients without imaging evidence of distant metastasis, diagnostic staging laparoscopy (DSL) is recommended to detect radiographically occult peritoneal metastasis (M1). DSL carries a risk for morbidity and its cost-effectiveness is unclear. Use of endoscopic ultrasound (EUS) to improve patient selection for DSL has been proposed but not validated. We aimed to validate an EUS-based risk classification system predicting risk for M1 disease. We retrospectively identified all GC patients without positron emission tomography (PET)/computed tomography (CT) evidence of distant metastasis who underwent staging EUS followed by DSL between 2010 and 2020. T1-2, N0 disease was EUS "low-risk"; T3-4 and/or N+ disease was "high-risk". A total of 68 patients met inclusion criteria. DSL identified radiographically occult M1 disease in 17 patients (25%). Most patients had EUS T3 tumors (n = 59, 87%) and 48 (71%) patients were node-positive (N+). Five (7%) patients were classified EUS "low-risk" and 63 (93%) were classified "high-risk". Of 63 "high-risk" patients, 17 (27%) had M1 disease. The ability of "low-risk" EUS to predict M0 disease at laparoscopy was 100% and DSL would have been avoided in five patients (7%). This stratification algorithm showed a sensitivity of 100% (95% confidence interval (CI): 80.5-100%) and a specificity of 9.8% (95% CI: 3.3-21.4%). Use of an EUS-based risk classification system in GC patients without imaging evidence of metastasis helps identify a subset of patients at low-risk for laparoscopic M1 disease who may avoid DSL and proceed directly to neoadjuvant chemotherapy or resection with curative intent. Larger, prospective studies are needed to validate these findings.