A New Clinical Model for Prediction of Nodal Metastasis for Patients with Early Gastric Cancer: Is this Good Enough?

Abstract

Early gastric cancer (EGC), an entity that was first recognized in 1962 by the Japanese Society of Gastroenterological Endoscopy, refers specifically to gastric cancer that is confined to the mucosa (T1a) or submucosa (T1b) layer irrespective of lymph node involvement. Such a distinction was established based on the observation that patients with EGC have an excellent prognosis after conventional gastrectomy with standardized lymph node dissection. Although nodal status does not affect the designation of EGC, lymph node metastasis is one of the most important prognostic factor. As a result, gastrectomy with regional lymphadenectomy remained the gold standard for treatment of EGC. However, radical gastrectomy with standard D2 nodal dissection is associated with significant risks of morbidity and mortality, and further compromises patients’ quality of life. Endoscopic resection emerged as a highly attractive alternative, because it preserves the stomach and averts all the complications associated with gastrectomy. Because endoscopic resection provides only local treatment without lymph node dissection, it should only be performed for EGC with absence of nodal metastases. The overall incidence of lymph node metastases in EGC ranged from 0 to 15 %; the rate was 0–3 % for intramucosal carcinomas and 11–20 % for submucosal tumors. Conventionally, the absolute indication for endoscopic resection of EGC was a well-differentiated, nonulcerative intramucosal carcinoma of less than 2 cm according to the Japanese guidelines. Gotoda et al. reviewed the nodal metastasis from gastrectomy specimens for patients with EGCs and identified a subgroup of patients with nonulcerative, intramucosal, well-differentiated EGC who have virtually no risk of nodal metastases. Consequently, the criteria for endoscopic resection was expanded (Table 1) as included in the current Japanese gastric cancer treatment guidelines to allow endoscopic resection of well-differentiated intramucosal EGC of any size. Although patients with nonulcerative superficial submucosal EGC of less than 3 cm were considered as adequately treated by endoscopic resection under the expanded criteria, there was still a low but definite risk of nodal metastasis. The clinical issue is how to better predict the risk of nodal metastasis for patients with EGC. Fujikawa and colleagues attempted to answer this question by coming up with a new prediction model of nodal status in patients with EGC—one that relies solely on clinical parameters, thereby improving its utility and practicality among clinicians. Of their 511 patients with T1 gastric cancer who did not meet the absolute criteria for endoscopic resection and thus received radical gastrectomy, 465 were found on pathology to have N0 disease. This suggested that radical gastrectomy was overtreating 91 % of their cohort of patients. Upon multivariate analysis, tumor diameter, depth of invasion, and histological type were significant independent clinical risk factors for nodal metastasis. These parameters were combined to calculate Fujikawa’s prediction model (Table 2). This model was able to further identify more than two-thirds of EGC patients without nodal metastasis who do not fall into the absolute criteria for endoscopic resection according to the Japanese guidelines. The ability of Fujikawa’s predictive criteria to identify new EGC candidates for endoscopic treatment based solely on clinical factors that can be obtained before treatment is very appealing. However, the caveats to reproducing these results depended heavily on Society of Surgical Oncology 2015