Liver metastasis-specific mortality in patients with primary gallbladder carcinoma: a competing risk model analysis...

To investigate the independent risk factors for liver metastasis-specific mortality in patients with gallbladder carcinoma and also to develop a competing risk model.

Patients diagnosed with gallbladder carcinoma between 2000 and 2019 in the Surveillance, Epidemiology, and End Results (SEER) database were included in this study. Patients' death of liver metastases was regarded as an outcome event and death of other causes as a competing event. Feature selection was performed using univariate and multivariate Fine & Gray regression. The area under the receiver operating character (ROC) curve (AUC) was calculated, and calibration curves were plotted to assess the performance of the constructed predictive model.

The study ultimately enrolled 8964 patients, with 6266 in the training cohort and 2698 in the validation cohort. The median follow-up duration for the total study cohort was 12 months, with a total of 1105 patients who died of liver metastases. The results of feature selection showed that histological type, median of household income, SEER stage, tumor sequence, sex, tumor size, and whether surgery was performed (Surgery) were independent factors influencing the liver metastasis-specific mortality in patients with gallbladder carcinoma. Result of calibration curve analysis showed that the model performed well and did not significantly overestimate or underestimate the risk of death.

The predictive model we constructed can help surgeons assess the risk of liver metastasis-specific death in patients with gallbladder carcinoma and help clinicians predict the prognosis of patients and make more rational clinical decisions.