To construct a nomogram to predict the 1-,3- and 5-year overall survival (OS) of colorectal cancer liver metastases(CRLM) patients, and to validate the prediction efficiency and clinical benefit of the model.

The clinical data of 113 colorectal liver metastasis patients who underwent resection of liver metastases in the Affiliated Hospital of Qingdao University from January 2008 to January 2022 were retrospectively analyzed. Using R software to implement random sampling, ninty-one patients (4/5) were selected as the training group, and the remaining 22 patients (1/5) were selected as the validation group. In the training group, use the multivariate Cox stepwise regression analysis to select 10 risk factors in the best survival prediction model. A nomogram for predicting 1-,3-and 5-year OS was established based on these risk factors. The nomogram's prediction efficiency was evaluated by concordance index (C-index), receiver operating characteristic curve (ROC) and calibration curve. The clinical benefit and application value of the model were evaluated by clinical decision curve analysis.

The results of multiple COX stepwise regression analysis based on training group showed that: age, BMI, the primary tumor histological grade, preoperative plasma CEA level, preoperative neutrophils-lymphocyte ratio (NLR), surgical margin, measured blood loss, postoperative targeted therapy, CRS score and postoperative recurrence and metastasis were the best risk factors for predicting the prognosis of CRLM patients. A nomogram was established and validated based on the risk factors. The C-index of OS in training and validation groups were 0.742(95%CI: 0.652~0.832) and 0.653(95%CI: 0.423~0.883). The average area under the curve (AUC) of ROC curves of 1-,3- and 5-year OS in the training and validation groups were all more than 0.7. Calibration curves also showed an excellent agreement between actual survival and nomogram predictive survival. The clinical decision curve analysis showed that model can bring higher clinical benefit.

The established nomogram can accurately predict the 1-,3-and 5-year OS of CRLM patients after operation. It may be helpful for clinical oncologists to evaluate the prognosis of CRLM patients and optimize diagnosis and treatment plan.