A clinical tool for early mortality risk stratification in colorectal cancer patients with liver or lung metastases: a retrospective cohort study

Background Colorectal cancer is the third most common cancer worldwide. Its insidious onset often leads to late diagnosis and tumor metastasis. Liver and lung are the most common sites of metastasis in patients with advanced colorectal cancer. Predicting the risk of early death in patients with metastatic colorectal cancer (mCRC) is essential. Method Univariate and multivariate logistic regression analyses were applied to identify independent variables affecting 6-month early mortality (defined as cancer-specific survival within 6 months of initial diagnosis) in mCRC patients. Variables were used to construct a nomogram to predict the probability of 6-month early mortality in mCRC patients. Receiver operating characteristic curve (ROC) curves and calibration curves were plotted to validate the predictive accuracy of the nomogram. Decision curve analysis (DCA) curves were plotted to evaluate the degree of benefit to the patients. Kaplan-Meier survival curves were used to demonstrate the differences in risk for the different patient strata. Result Age, surgery, radiotherapy, chemotherapy, tumor grade, T stage, N stage and metastasis status were identified as independent factors affecting the early prognosis of mCRC patients who developed hepatic or pulmonary mCRC and were used to construct the nomogram. Good results were obtained for the area under curve (AUC) of both training cohort and validation cohort (0.851 and 0.845). The calibration curves were close to 45° demonstrating the good accuracy of the nomogram, and the DCA curves and the Kaplan-Meier survival curves demonstrated that the nomogram could provide good results for the patients. Conclusion We have successfully developed a nomogram that accurately predicts the probability of early death in mCRC patients with liver or lung metastases, and this nomogram has been validated and shown to have excellent predictive ability. This can effectively help clinicians to choose better treatment strategies for these patients.