Analysis of Prognostic Risk Factors and Establishment of a Prognostic Model for Liver Metastasis in Patients with Colon Cancer

Abstract

Abstract Background Colorectal cancer often metastasizes to the liver, which is associated with poor prognosis. The aim of this study was to establish an efficient nomogram model for predicting overall survival (OS) and disease-free-survival (DFS) in patients with colorectal cancer and liver metastasis. Methods We analyzed 421 patients diagnosed with colorectal cancer and liver metastasis at Tongji Medical College of Huazhong University of Science and Technology and Affiliated Union Hospital from January 2013 to December 2018. These patients were randomly assigned to training and validation cohorts. Single-factor and multivariate Cox regression analyses were performed to determine independent predictive risk factors and to construct nomograms for predicting OS and DFS. The performance of the nomograms was evaluated using calibration curves, area under the receiver operating characteristic curve (AUROC), and decision curve analysis (DCA). Results Tumor size, vascular tumor embolus, blood transfusion, number of liver metastases, number of sampled lymph nodes, staging, postoperative hospital stay, and carcinoembryonic antigen (CEA) were identified as independent predictive factors for liver metastasis. We developed a survival prediction model incorporating these eight prognostic factors. The Nomogram demonstrated good sensitivity in predicting 1-year, 3-year, and 5-year OS rates. In the training cohort, the AUROC for 1-year, 3-year, and 5-year OS was 0.793, 0.758, and 0.823, respectively. In the validation cohort, the respective AUROC values were 0.750, 0.704, and 0.822, respectively. Additionally, we constructed a column chart for patients' DFS based on histological subtype, number of sampled lymph nodes, vascular tumor embolus, number of liver metastases, perioperative transfusion, and CEA level. In the training cohort, the 1-year, 3-year, and 5-year DFS rates were 0.768, 0.716, and 0.803, respectively. In the validation cohort, the rates were 0.730, 0.839, and 0.838, respectively. Conclusion Based on clinical, pathological, and tumor biomarker characteristics, the newly constructed nomogram accurately predicted OS and DFS. This tool may be valuable for guiding clinical decision-making. In practice, individual patient data and analytical results may be used to develop personalized treatment plans that may improve prognosis and overall survival rates.