RESUMO
Objective:To identify the independent risk factors affecting the prognosis of gallbladder cancer after radical resection, and to develop and validate the nomogram of predictive model.Methods:The clinical data of 147 patients with gallbladder cancer treated in the First Affiliated Hospital of Xinjiang Medical University from January 2012 to January 2022 were retrospectively analyzed, including 53 males and 94 females, aged (61.45±10.76) years old. The patients were followed up by outpatient or telephone review. The Kaplan-Meier method and log-rank test were used for survival analysis. The variables of P<0.1 in univariate analysis were included in the minimum absolute convergence and selection operator (LASSO) regression model, and the predictive factors affecting the prognosis of gallbladder cancer were screened. The predictive model was established by multivariate Cox regression analysis, and a nanogram was constructed based on the multivariate Cox regression model. The discrimination of the model was evaluated by consistency index (C index), time-dependent C index curve, receiver operating characteristic curve and area under the curve (AUC). 500 times of Bootstrap sampling were conducted for the calibration of nomogram. Results:The median survival of patients with gallbladder cancer was 22.15 months, and the 1-, 2- and 3-year cumulative survival rates were 65.99%, 46.02% and 35.73%, respectively. LASSO regression analysis showed that age, abdominal pain, degree of differentiation, T stage, N stage, serum levels of CA-199 and total bilirubin were predictive factors affecting the prognosis of gallbladder cancer (all P<0.05). The prognosis prediction model was established by multivariate Cox regression analysis. The C-index was 0.856 (95% CI: 0.823-0.887). The AUC values for 1-year and 3-year survival probabilities are 0.939 and 0.944, respectively. The calibration chart indicates that this model has a good accuracy. The decision curve analysis confirmed that the net benefit of this model is significantly higher than two extreme situations, indicating its clinical applicability and patients’ benefits. Conclusion:The nomogram for postoperative prognosis of gallbladder cancer based on age, abdominal pain, degree of differentiation, T stage, N stage, serum levels of total bilirubin and CA-199 has a high accuracy, which might affect the treatment decision-making of patients with gallbladder cancer.