Preoperative MRI-based nomogram to predict survival after curative resection in patients with gallbladder cancer: a retrospective multicenter analysis.

ABSTRACT

PURPOSE: To use preoperative MRI data to construct a nomogram to predict survival in patients who have undergone R0 resection for gallbladder cancer. METHODS: The present retrospective study included 143 patients (MF, 7667; 67.15 years) with gallbladder cancer who underwent preoperative MRI and subsequent R0 resection between 2013 and 2021 at two tertiary institutions. Clinical and radiological features were analyzed using univariate and multivariate Cox regression analysis to identify independent prognostic factors. Based on the multivariate analysis, we developed an MRI-based nomogram for determining prognoses after curative resections of gallbladder cancer. We also obtained calibration curves for 1-,3-, and 5-year survival probabilities. RESULTS: The multivariate model consisted of the following independent predictors of poor overall survival (OS), which were used for constructing the nomogram age (years; hazard ratio [HR] = 1.04; 95% confidence interval [CI], 1.04-1.07; p = 0.033); tumor size (cm; HR = 1.40; 95% CI, 1.09-1.79; p = 0.008); bile duct invasion (HR = 3.54; 95% CI, 1.66-7.58; p = 0.001); regional lymph node metastasis (HR = 2.47; 95% CI, 1.10-5.57; p = 0.029); and hepatic artery invasion (HR = 2.66; 95% CI, 1.04-6.83; p = 0.042). The nomogram showed good probabilities of survival on the calibration curves, and the concordance index of the model for predicting overall survival (OS) was 0.779. CONCLUSION: Preoperative MRI findings could be used to determine the prognosis of gallbladder cancer, and the MRI-based nomogram accurately predicted OS in patients with gallbladder cancer who underwent curative resection.