Development and External Validation of a Model to Predict Overall Survival in Patients With Resected Gallbladder
Elise A J de Savornin Lohman1,2, T J J de Bitter3, G Hannink4
1Radboud University Medical Center, Radboud Institute for Health Sciences, Department of Surgery, Nijmegen, The Netherlands.
Annals of Surgery
|August 13, 2021
Summary
A new clinical prediction model accurately forecasts survival for resected gallbladder cancer (GBC) patients. This model outperforms the AJCC staging system, offering improved prognostic insights for GBC survival.
Area of Science:
- Oncology
- Surgical Oncology
- Biostatistics
Background:
- Assessing survival in resected gallbladder cancer (GBC) remains challenging.
- Existing tools lack optimal accuracy for predicting outcomes in GBC patients.
Purpose of the Study:
- To develop and validate a clinical prediction model for overall survival in nonmetastatic, resected GBC.
- To compare the model's performance against the established American Joint Committee on Cancer (AJCC) staging system.
Main Methods:
- Utilized a Dutch nationwide cohort for model development (380 patients).
- Performed internal validation and external validation using an Australian cohort (66 patients).
- Compared the developed model with the AJCC staging system using C-index metrics.
Main Results:
- Identified age, T/N classification, resection margin, differentiation grade, and vascular invasion as independent predictors.
- The model demonstrated strong discriminatory capacity with an external validation C-index of 0.75.
- The developed model significantly outperformed the AJCC staging system (0.71 vs. 0.59).
Conclusions:
- The novel prediction model offers good discriminatory capacity and calibration for resected GBC survival.
- This tool surpasses the AJCC staging system, providing valuable prognostic information.
- The model is accessible online for clinical use by physicians and patients.
