Potential Advantages of Robotic Total Gastrectomy in Advanced Middle-Upper Gastric Cancer: A Multicenter Propensity Score Matching Analysis

All patients provided informed consent for the use of their clinical data at admission or prior to surgery, in accordance with the Declaration of Helsinki. The study protocol was approved by the institutional review boards of all eight participating centers and reported in accordance with STROCSS guidelines.

This multicenter registry included patients diagnosed with middle and upper gastric cancer (GC), who underwent radical total gastrectomy with pathological staging of T2-4aN0-3bM0. Exclusion criteria included ASA class >3, residual GC, neoadjuvant chemotherapy, combined organ resection, indocyanine green (ICG) use, concurrent/past malignancies, and loss to follow-up. A total of 1,099 eligible patients were included in the final analysis, with 237 receiving robotic total gastrectomy (RTG) and 862 receiving laparoscopic total gastrectomy (LTG).

To ensure surgical quality, all centers followed standardized protocols. Surgeons had completed their institutional learning curve-defined as independently performing ≥50 laparoscopic gastrectomies with consistent outcomes-before enrolling patients. RTG was performed only by certified surgeons trained in the Da Vinci Robotic System. Surgical procedures adhered to Japanese Gastric Cancer Treatment Guidelines, and pathological staging followed the AJCC 8th edition criteria. Resected specimens were evaluated by experienced gastrointestinal pathologists using uniform histopathological protocols. R1 resection was defined as microscopic tumor presence at the surgical margin.

Patient selection for RTG or LTG was based on shared decision-making, considering tumor stage, comorbidities, surgical risk, cost, and patient preference. Informed consent included agreement to cover any additional costs associated with robotic surgery.

Propensity score matching (PSM) was used to control for baseline differences between groups. A 1:1 nearest-neighbor matching with a caliper of 0.2 was performed using the MatchIt package in R. Thirteen covariates-age, sex, BMI, ASA grade, comorbidities, tumor size/location, histology, pT/pN stage, lymphadenectomy extent, adjuvant chemotherapy, and year of surgery-were included. Matching balance was evaluated using standardized mean differences (SMDs), with <0.1 considered acceptable.

The registry incorporated a structured follow-up system. Patients were followed every 3-6 months in the first two years, every 6-12 months between years 3-5, and annually thereafter. Follow-up assessments included physical exams, lab tests, imaging (X-ray, ultrasound, or CT), and annual endoscopy. Telephone follow-up was used to supplement outpatient visits. Recurrence was confirmed through clinical history, imaging, and pathology (preferably biopsy).

Data integrity was ensured through multiple layers of quality control. All participating centers used a standardized electronic data capture (EDC) system. Data entry was performed by trained personnel and verified by designated staff. Regular audits, backtracking of missing data, and cross-checking by two independent researchers were conducted before data export. These procedures ensured accuracy, completeness, and consistency of the registry data.

Statistical analyses were performed using SPSS v26.0 and R v4.4.1. Continuous variables were analyzed via t-tests or Mann-Whitney U tests; categorical variables via chi-square or Fisher's exact tests. Survival was assessed using Kaplan-Meier and Cox regression models, with competing risk models used for recurrence. All tests were two-sided, with a significance level of P<0.05.