Preoperative nomogram to predict survival following colorectal cancer liver metastasis simultaneous resection.

backgroundSimultaneous resection for patients with synchronous colorectal cancer liver metastases (CRLM) remains an optimal option for the sake of curability. However, few studies so far focus on outcome of this subgroup of patients (who receive simultaneous resection for CRLM). Substantial heterogeneity exists among such patients and more precise categorization is needed preoperatively to identify those who may benefit more from surgery. In this study, we formulated this internally validated scoring system as an option.

methodsClinicopathological and follow-up data of 234 eligible CRLM patients undergoing simultaneous resection from January 2010 to March 2019 in our center were included for analysis. Patients were randomized to either a training or validation cohort. We performed multivariable Cox regression analysis to determine preoperative factors with prognostic significance using data in training cohort, and a nomogram scoring system was thus established. Time-dependent receiver operating characteristic (ROC) curve and calibration plot were adopted to evaluate the predictive power of our risk model.

resultsIn the multivariable Cox regression analysis, five factors including presence of node-positive primary defined by enhanced CT/MR, preoperative CEA level, primary tumor location, tumor grade and number of liver metastases were identified as independent prognostic indicators of overall survival (OS) and adopted to formulate the nomogram. In the training cohort, calibration plot graphically showed good fitness between estimated and actual 1- and 3-year OS. Time-dependent ROC curve by Kaplan-Meier method showed that our nomogram model was superior to widely used

conclusionsIn this retrospective analysis we identified several preoperative factors affecting survival of synchronous CRLM patients undergoing simultaneous resection. We also constructed and validated a risk model which showed high accuracy in predicting 1- and 3-year survival after surgery. Our risk model is expected to serve as a predictive tool for CRLM patients receiving simultaneous resection and assist physicians to make treatment decision.