Abstract:Background: The occurrence of liver metastasis significantly affects the prognosis of colorectal cancer (CRC). Existing research indicates that primary tumor location, vascular invasion, lymph node metastasis, and abnormal preoperative tumor markers are risk factors for CRC liver metastasis. Positive expression of programmed cell death ligand 1 (PD-L1) may serve as a favorable prognostic marker for nasopharyngeal and gastric cancers, in which combined positive score (CPS) quantifies the level of PD-L1 expression. This study aimed to explore CPS as a potential risk factor for CRC liver metastasis and integrate other independent risk factors to establish a novel predictive model for CRC liver metastasis. Methods: A retrospective analysis was conducted on 437 patients with CRC pathologically diagnosed at The Second Xiangya Hospital of Central South University from January 1, 2019, to December 31, 2021. Data were collected, including CPS, age, gender (male and female), primary tumor location, Ki-67 expression, pathologic differentiation, neural invasion, vascular invasion, lymph node metastasis, and preoperative tumor markers. The optimal cutoff point for the continuous variable CPS was determined using the Youden index, and all CPSs were dichotomized into high- and low-risk groups based on this threshold (scores below the threshold were considered high risk, and score above the threshold were considered low risk). Univariate logistic regression analysis was employed to identify risk factors for CRC liver metastasis, followed by multivariate logistic regression analysis to integrate the selected risk factors. The predictive model was validated through the construction of receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA). A nomogram was constructed for visualization. Results: The determined cutoff point for PD-L1 CPS was 4.5, with scores below this threshold indicating a high risk of CRC liver metastasis. In addition, primary tumor origin other than the rectum, presence of pericolonic lymph node metastasis, and abnormal levels of tumor markers carcinoembryonic antigen and cancer antigen 19-9 were identified as independent risk factors for CRC liver metastasis. The constructed clinical prediction model demonstrated good predictive ability and accuracy, with an area under the ROC curve of 0.871 (95% CI, 0.838-0.904). Conclusion: The exploration and validation of CPS as a novel predictor of CRC liver metastasis were performed. Based on these findings, a new clinical prediction model for CRC liver metastasis was developed by integrating other independent risk factors. The DCA, clinical impact curve, and nomogram graph constructed on the basis of this model have significant clinical implications and guide clinical practice.

Predicting biomarkers from classifier for liver metastasis of colorectal adenocarcinomas using machine learning models

Identification of novel diagnostic biomarkers associated with liver metastasis in colon adenocarcinoma by machine learning

Enhancing metastatic colorectal cancer prediction through advanced feature selection and machine learning techniques

Application of machine learning techniques in real-world research to predict the risk of liver metastasis in rectal cancer

Machine learning for predicting liver and/or lung metastasis in colorectal cancer: a retrospective study based on the SEER database

Using machine learning approach for screening metastatic biomarkers in colorectal cancer and predictive modeling with experimental validation

Assessment of Primary Colorectal Cancer CT Radiomics to Predict Metachronous Liver Metastasis

Comparison of AdaBoost and Logistic Regression for Detecting Colorectal Cancer Patients with Synchronous Liver Metastasis

Utilizing machine learning algorithms for predicting risk factors for bone metastasis from right-sided colon carcinoma after complete mesocolic excision: a 10-year retrospective multicenter study

Colon cancer diagnosis and staging classification based on machine learning and bioinformatics analysis

Rapid detection of liver metastasis risk in colorectal cancer patients through blood test indicators

Establishment of a New Non-Invasive Imaging Prediction Model for Liver Metastasis in Colon Cancer

Development of a Prediction Model for Lung Metastasis by Using Postoperative Circulating Tumor DNA Binding Clinicopathological Risk Factors in Patients with Colorectal Cancer.

Support vector machine classifier for prediction of the metastasis of colorectal cancer

Application of machine learning algorithm in prediction of lymph node metastasis in patients with intermediate and high-risk prostate cancer

Deep Learning-Based Prediction of Future Extrahepatic Metastasis and Macrovascular Invasion in Hepatocellular Carcinoma

Risk Assessment of Liver Metastasis in Pancreatic Cancer Patients Using Multiple Models Based on Machine Learning: A Large Population-Based Study

Enhancing the diagnostic accuracy of colorectal cancer through the integration of serum tumor markers and hematological indicators with machine learning algorithms

Predictive modeling algorithms for liver metastasis in colorectal cancer: A systematic review of the current literature

Identification of Metastasis-Related Genes for Predicting Prostate Cancer Diagnosis, Metastasis and Immunotherapy Drug Candidates Using Machine Learning Approaches

Multifactorial risk prediction analysis of liver metastasis in colorectal cancer: incorporating programmed cell death ligand 1 combined positive score and other factors