Fig. 6
Performance of the 3-marker methylation model and integrated model as compared to the standard workups. a Treatmental modalities of EGC patients in the two cohorts as guided by the 3-marker methylation model and the integrated model as compared to standard workups (JGCA). Absolute indication for endoscopic resection (ERI), differentiated EGC confined to the mucosa without ulcerative findings and with a tumor diameter ≤ 2 cm. Expanded indication for endoscopic resection, mucosal gastric cancers of the following categories: (i) differentiated-type, ulceration (absence), diameter > 2 cm; (ii) differentiated-type, ulceration (presence), diameter ≤ 3 cm; and (iii) undifferentiated-type, ulceration (absence), diameter ≤ 2 cm). Relative indication for endoscopic resection, EGC tumors that do not fulfill the absolute or expanded indications. b–d The undertreatment rate, overtreatment rate, and accuracy of the 3-marker methylation model and the integrated model compared to standard workups in patients as indicated for absolute indication, expanded indication, and the relative indication of endoscopic resection (n = 29, 81 and 322). e A overall estimation of undertreatment rate, overtreatment rate, and accuracy analysis of the 3-marker model and the integrated model as compared to the standard workups. Statistical significance was assessed using χ2 test. *p < 0.05, **p < 0.01, ***p < 0.001, NS., not statistically significant