Endoscopic resection for undifferentiated early gastric cancer: focusing on histologic discrepancies between forceps biopsy-based and endoscopic resection specimen-based diagnosis.

Before endoscopic resection (ER), a considerable number of undifferentiated early gastric cancer (UD-EGC) cases were initially diagnosed as atypical glands, dysplasia, or differentiated EGC (D-EGC) based on forceps biopsy specimens. As UD-EGC carries a high risk of resection margin involvement, identifying the predictive factors for UD-EGC cases with histologic discrepancy (HD) is of clinical importance. To investigate the outcomes of ER for UD-EGC and to identify the predictive factors for UD-EGC with HD. Among 2,194 EGC lesions treated by ER, 59 lesions were finally diagnosed as UD-EGC and 50 UD-EGC cases showed HD. The demographic and endoscopic characteristics were compared between D-EGC and UD-EGC with HD, and the predictive factors for the latter were investigated among cases of forceps biopsy-based diagnosis of atypical glands, dysplasia, or D-EGC. UD-EGC showed significantly higher rate of lateral margin involvement compared to D-EGC (18.6 vs. 3.4%). Among the UD-EGC cases meeting the expanded criteria and not involving additional surgery, no local or extragastric tumor recurrence was observed during the median follow-up of 27.5 months. Multivariate analysis demonstrated that age (≤60 years), female gender, gastric body, flat or depressed type, and tumor size (>2 cm) were independent predictive factors for UD-EGC with HD among cases of forceps biopsy-based diagnosis of atypical glands, dysplasia, or D-EGC. For lesions with predictive factors for UD-EGC with HD, a circumferential mapping biopsy before ER or wide marking during ER could be considered to avoid the potential risk of incomplete resection.