Abstract
BACKGROUND: The optimal surgical strategy for early-stage gallbladder cancer (GBC) remains controversial. This study aimed to compare the long-term survival outcomes of simple cholecystectomy (SC) and extended cholecystectomy (EC) in patients with stage I GBC. METHODS: Patients diagnosed with stage I GBC between 2010 and 2015 were identified from the Surveillance, Epidemiology, and End Results (SEER) database and categorized into SC and EC groups. Kaplan-Meier survival analyses, propensity score matching (PSM), and competing risk models were applied. Competing risk nomograms were subsequently constructed and externally validated. RESULTS: A total of 902 patients were included, along with 33 patients from Hangzhou TCM Hospital as the external validation cohort. After 1:1 PSM, the EC group demonstrated lower cancer-specific death (CSD) rates at 1, 3, and 5 years compared with the SC group (1-year: 6.43% vs. 11.5%; 3-year: 12.5% vs. 24.4%; 5-year: 15.3% vs. 28.9%; P = 0.002). Subgroup analyses indicated that SC was associated with increased CSD in patients with T1b disease, grade II-IV tumors, and adenocarcinoma. Nomograms incorporating grade, T stage, tumor size, and chemotherapy accurately predicted individualized 1-, 3-, and 5-year CSD probabilities. The models exhibited strong discriminative ability (AUC values) and good calibration performance. CONCLUSIONS: Competing risk analysis suggested that EC should be recommended for patients with T1b GBC, whereas SC alone may be sufficient for T1a disease. The proposed nomograms demonstrated robust predictive performance and may aid clinical decision-making.