Background: Coagulation and fibrinolysis activation are frequently observed in cancer patients, and the tumors in these cases are thought to be associated with a higher risk of invasion, metastasis and worse long-term outcome. The specific aim of this study was to develop an effective prognostic nomogram to help make individualized estimates for patients with resected gallbladder cancer (GBC).
Methods: Patients with resected GBC who were diagnosed between 2006 and 2014 at Xinhua Hospital were selected. Model performance was measured by c-index and calibration curve. The results were further validated using bootstrap and a cohort of 38 patients from a branch hospital who underwent surgery from 2006 to 2014.
Results: Backward stepwise selection and Lasso were applied respectively to select predictors. T stage, N stage, and preoperative serum fibrinogen were included in the final model. Predictions correlated well with observed 1- and 3-year survival. The c-index for predicting survival was 0.74 (95% confidence interval, 0.70-0.78), which was statistically higher than that of the AJCC 7th system and Nevin system (P=0.04, 0.04, respectively). In the validation cohort, the nomogram performed better than the other two staging systems (c-index: 0.71 vs. 0.67 and 0.67).
Conclusions: The validated nomogram is a practical tool for predicting the overall survival (OS) of postoperative GBC patients. Preoperative serum fibrinogen levels were associated with tumor progression and may be an independent predictor for GBC patients.
Keywords: Clinical prediction model; fibrinogen; gallbladder cancer (GBC); nomogram.
2021 Journal of Gastrointestinal Oncology. All rights reserved.