Objective To explore the risk factors for very early recurrence (VER) after curative-intent resection for gallbladder cancer (GBC) patients and construct prediction models for VER based on various machine learning (ML) algorithms. Methods A retrospective study was conducted on 329 GBC patients who underwent curative-intent surgery at our hospital between January 2016 and December 2020. Risk factors for VER were identified, and prediction models were constructed, validated and compared with multiple ML algorithms[logistic regression (LR), support vector machine (SVM), naive Bayes (NB), random forest (RF), light gradient boosting machine (LGB), and extreme gradient boosting (XGB)]based on independent associated factors for VER. Results Among the 329 patients who underwent curative-intent resection in patients with GBC, 162 (49.2%) patients experienced recurrence, including 69 (42.6%) with VER（<6 months） and 93 (57.4%) with non-VER（≥6 months）. Survival analysis showed that patients with VER had significantly worse median overall survival compared to those with non-VER (6 months vs. not arrived，χ2=398.2, P<0.001). Univariate analysis showed that carcinoembryonic antigen (CEA), carbohydrate antigen (CA)19-9, CA-125, tumor differentiation, pathological type, liver involvement, vascular invasion, perineural invasion, TNM stage, T stage and N stage were risk factors of VER (P<0.05), whereas adjuvant chemotherapy was protective factor (P<0.05). Multivariate analysis confirmed CA-125, tumor differentiation, pathological type, vascular invasion and N stage as independent risk factors (P<0.05), whereas adjuvant chemotherapy was independent protective factor (P<0.05). XGB model achieved the best performance with an area under curve (AUC) of 0.841 and an accuracy (ACC) of 83.0% in the validation set. Shapley additive explanations (SHAP) bar plots highlighted tumor differentiation, N stage, pathological type of tumor, and CA-125 the top four features contributing to the model, each positively influencing the predicted probability of VER. Conclusions CA-125, tumor differentiation, pathological type, vascular invasion, N stage and adjuvant chemotherapy are independent factors associated with VER of GBC following curative-intent resection. ML-based prediction models incorporating these factors have the potential to some extent to effectively identify high-risk patients, providing a valuable reference for VER surveillance in GBC.

Objective:To investigate the efficacy of liver wedge resection and liver Ⅳb and Ⅴ segmentectomy for T2 gallbladder carcinoma (GBC).Methods:The retrospective cohort study was conducted. The clinicopathological data of 168 patients who underwent radical resection of T2 GBC in The First Affiliated Hospital of Xi′an Jiaotong University from January 2011 to December 2021 were collected. There were 59 males and 109 females, aged (65±10)years. Of 168 patients, there were 112 cases in T2a stage and 56 cases in T2b stage. Of 112 patients in T2a stage, 73 cases underwent liver wedge resection and 39 cases underwent liver Ⅳb and Ⅴ segmentectomy. Of 56 patients in T2b stage, 27 cases underwent liver wedge resection and 29 cases underwent liver Ⅳb and Ⅴ segmen-tectomy. Measurement data with normal distribution were represented as Mean± SD, and measure-ment data with skewed distribution were represented as M(range). Count data were described as absolute numbers, and comparison between groups was conducted using the chi-square test or Fisher exact probability. Comparison of ordinal data was conducted using the Mann-Whitney U test. The Kaplan-Meier method was used to calculate survival rate and draw survival curve, and the Log-rank test was used for survival analysis. The COX proportional risk model was used for univariate and multivariate analyses. Results:(1) Clinical data analysis of patients undergoing different extent of hepatic resection for T2 GBC. There was no significant difference in gender, age, cholecystoli-thiasis, preoperative total bilirubin, carcinoembryonic antigen, CA19-9, CA125, incidental GBC, perineural invasion, microvascular invasion, pathological differentiation, histopathological subtypes, N staging, TNM staging between patients with T2a and T2b GBC who underwent different extent of hepatic resection ( P>0.05). (2) Prognostic analysis of T2 GBC patients undergoing different extent of hepatic resection. The 1-, 3- and 5-year cumulative disease-free survival rates of T2 GBC patients undergoing liver wedge resection were 78.0%, 60.1% and 51.4%, respectively, versus 86.8%, 80.0% and 68.0% of T2 GBC patients undergoing liver Ⅳb and Ⅴ segmentectomy, showing a significant difference between them ( χ2 =5.205, P<0.05). The 1-, 3-, and 5-year cumulative overall survival rates of T2 GBC patients undergoing liver wedge resection were 85.0%, 62.5%, and 55.1%, respectively, versus 92.6%, 81.6%, and 68.8% for T2 GBC patients undergoing liver Ⅳb and Ⅴ segmentectomy, showing a significant difference in cumulative overall survival rate between them ( χ2=4.351, P<0.05). The 1-, 3-, and 5-year cumulative disease-free survival rates of T2b GBC patients undergoing liver wedge resection were 70.4%, 45.9% and 39.2%, respectively, versus 89.7%, 71.3% and 54.0% of T2b GBC patients undergoing liver Ⅳb and Ⅴ segmentectomy, showing a significant difference between them ( χ2=5.047, P<0.05). The 1-, 3-, and 5-year cumulative overall survival rates of T2b GBC patients undergoing liver wedge resection were 81.5%, 53.2%, and 41.0%, respectively, versus 89.7%, 77.0%, and 60.7% of T2b GBC patients undergoing liver Ⅳb and Ⅴ segmentectomy, showing no significant difference in cumulative overall survival rate between them ( χ2=4.014, P<0.05). (3) Analysis of factors influencing prognosis of patients undergoing radical resection for T2 GBC. Results of multivariate analysis showed that CA19-9>39.0 U/mL, perineural invasion, N1 and N2 stage were independent risk factors influencing disease-free survival time of patients undergoing radical resection for T2 GBC ( hazard ratio=2.736, 3.496, 2.638, 17.440, 95% confidence interval as 1.195-6.266, 1.213-10.073, 1.429-4.869, 8.362-36.374, P<0.05). Liver Ⅳb and Ⅴ segmentectomy was an independent protective factor influencing disease-free survival time of patients undergoing radical resection for T2 GBC ( hazard ratio=0.418, 95% confidence interval as 0.230-0.759, P<0.05). CA19-9 >39.0 U/mL, perineural invasion, ⅡB stage, ⅢB stage and ⅣB stage of TNM staging were independent risk factors influencing overall survival time of patients undergoing radical resection for T2 GBC ( hazard ratio=2.740, 3.210, 2.037, 3.439, 24.466, 95% confidence interval as 1.127-6.664, 1.049-9.819, 1.004-4.125, 1.730-6.846, 10.733-55.842, P<0.05). Liver Ⅳb and Ⅴ segmentectomy was an independent protective factor influencing overall survival time of patients undergoing radical resec-tion for T2 GBC ( hazard ratio=0.476, 95% confidence interval as 0.261-0.867, P<0.05). (4) Analysis of postoperative complications in patients undergoing different extent of hepatic resection for T2 GBC. There was no significant difference in postoperative complications of patients with T2a and T2b GBC undergoing liver wedge resection or liver Ⅳb and Ⅴ segmentectomy ( P>0.05). Conclusions:Compared to liver wedge resection, liver Ⅳb and Ⅴ segmentectomy can effectively prolong the disease-free survival overall survival time of T2b GBC patients. There is no significant difference in the major complications. Liver Ⅳb and Ⅴ segmentectomy is an independent protective factor for prognosis of patients undergoing radical resection for T2 GBC.