Histological Mixed-Type Early Gastric Cancer Is an Independent Risk Factor for Lymph Node Metastasis: Presidential Poster

Abstract

Introduction: Endoscopic submucosal dissection (ESD) is a promising method for early gastric cancer (EGC). In certain ESD cases, additional surgery is to be considered after the histological investigation of the obtained specimen. However, management of lesions which have differentiated and undifferentiated carcinoma (mixed-type) is still unclear. The aim of the study is to clarify the clinicopathological significance of histological mixed-type EGC. Methods: We retrospectively reviewed 593 consecutive EGC lesions in 528 subjects treated in our hospital (age (mean±SE) 69.9±0.4), male/female 410/183). 362 were underwent ESD and 231 were underwent gastrectomy. Histological findings of the lesions were classified into following four groups; purely undifferentiated type (PD), mixed-type with predominantly differentiated part (MD), purely undifferentiated type (PU), and mixed-type with predominantly undifferentiated part (MU). We investigated the correlation between these types and following findings; location of the tumor, macroscopic type, tumor size, presence of lymphatic and vascular duct invasion, submucosal invasion depth, and the prevalence of lymph node metastasis. Among subjects underwent gastrectomy, potential risk factor for lymph node metastasis including macroscopic type, tumor size, submucosal invasion depth, and the histological type (pure/mixed) were evaluated by multivariate logistic regression analysis. Results: Subjects with PU type cancer were younger and held more female subjects than those with MU type (p < 0.001, p < 0.05). In macroscopic type, MD and PU type included more depressed lesions than PD and MU type, respectively (p < 0.0001, p < 0.05). Diameter of lesions of MD type were larger than that of PD type (p < 0.0001). MD and MU type were significantly associated with submucosal invasion than PD and PU type, respectively (p < 0.0001, p < 0.001) and this association was observed even among the lesion less than 20mm, respectively (p < 0.001, p < 0.05). Multivariate analysis revealed that mixed-type (odds ratio (OR) =3.1, 95% confidence interval (95%CI):1.2-8.3, p=0.02) and the invasion depth (OR=5.2, 95%CI: 1.6-2.3, p=0.01) were significant risk factor for lymph node metastasis. Conclusion: Histological mixed-type EGC is an independent risk factor for lymph node metastasis regardless of tumor size. When ESD is performed for EGC, histological confirmation of the resected tumor especially for pure or mixed type should be considered.