Objective This study sought to evaluate the effectiveness of the w-eCura system in determining the necessity of salvage surgery after non-radical endoscopic submucosal dissection (ESD) for early gastric cancer (EGC) and to compare its efficacy with the eCura system. Methods This retrospective case-control study was conducted at Fujian Provincial Hospital and included 290 patients with EGC who underwent ESD between January 2011 and July 2024. All patients were confirmed with non-radical resection upon pathological evaluation. Various clinicopathological characteristics such as gender, age, lesion location, size, Paris classification, histological type, depth of invasion, ulceration, lymphatic and venous invasion, and margin status were documented. Subsequent salvage procedures, lymph node involvement, and recurrence were monitored. Risk stratification was performed using the w-eCura and eCura systems, and the Cox proportional hazards model was employed to identify factors influencing positive cancer residue status (CRS). Survival analysis was conducted using Kaplan-Meier curves, and the predictive performance of the two models was compared using receiver operating characteristic (ROC) curves. Results The study included patients with a mean age of 64.42 ± 9.35 years and a median maximum tumor diameter of 4.50 cm, who were monitored for a median duration of 32 (range 15–54) months. Factors such as w-eCura grade, horizontal resection margin, depth of invasion, lymphatic invasion, and venous invasion exhibited significant associations with CRS-positive classification (P < .05). A higher w-eCura grade was linked to an elevated risk of CRS-positive classification. Utilizing Cox proportional hazard models, we identified that lesion location (The middle 1/3 stomach), w-eCura grade (high and intermediate), and infiltration depth (SM2 and SM3) independently influenced CRS-positive classification. Kaplan-Meier survival analysis demonstrated a significant correlation between both eCura and w-eCura grades and survival time (Logrank P < .001). Notably, the w-eCura classification displayed enhanced sensitivity in detecting high-risk patients promptly. ROC curve analysis revealed that the w-eCura model exhibited an area under the curve (AUC) of 0.82, surpassing the eCura model’s AUC of 0.79 and underscoring the superior predictive accuracy of the w-eCura model for CRS-positive classification. Within the intermediate-risk w-eCura subgroup, positive vertical resection margin and SM3 infiltration depth emerged as independent factors influencing CRS-positive classification. Conclusion The w-eCura system demonstrated superiority as a tool for assessing the necessity of salvage surgery following non-curative ESD in EGC compared to the eCura system. For patients presenting intermediate-risk, the w-eCura system in conjunction with positive vertical resection margin or SM3 depth of invasion, additional salvage surgery is advised for consideration.
Chen et al. (Tue,) studied this question.