Nomogram for predicting occurrence of synchronous liver metastasis in colorectal cancer: a single-center...

The purpose of this study was to explore the risk factors for synchronous liver metastasis (LM) of colorectal cancer (CRC) and to construct a nomogram for predicting the occurrence of synchronous LM based on baseline and pathological information.

The baseline and pathological information of 3190 CRC patients were enrolled in the study from the Department of Colorectal Surgery, the Second Affiliated Hospital of Harbin Medical University between 2012 and 2020. All patients were divided into development and validation cohorts with the 1:1 ratio. The characters of LM and none-LM patients in newly diagnosed colorectal cancer were utilized to explore the risk factors for synchronous LM with the univariate and multivariate logistic regression analyses. A predictive nomogram was constructed by using an R tool. In addition, receiver operating characteristic (ROC) curves was calculated to describe the discriminability of the nomogram. A calibration curve was plotted to compare the predicted and observed results of the nomogram. Decision-making curve analysis (DCA) was used to evaluate the clinical effect of nomogram.

The nomogram consisted of six features including tumor site, vascular invasion (VI), T stage, N stage, preoperative CEA, and CA-199 level. ROC curves for the LM nomogram indicated good discrimination in the development (AUC = 0.885, 95% CI 0.854-0.916) and validation cohort (AUC = 0.857, 95% CI 0.821-0.893). The calibration curve showed that the prediction results of the nomogram were in good agreement with the actual observation results. Moreover, the DCA curves determined the clinical application value of predictive nomogram.

The pathologic-based nomogram could help clinicians to predict the occurrence of synchronous LM in postoperative CRC patients and provide a reference to perform appropriate metastatic screening plans and rational therapeutic options for the special population.