Construction of a preoperative lymph node metastasis risk prediction model for colorectal cancer

Objective To investigate the independent risk factors among preoperative systemic inflammatory indicators associated with lymph node metastasis (LNM) in patients with colorectal cancer (CRC) and to construct and validate a related risk prediction model. Methods Clinical data of 241 patients with CRC who received surgery at Affiliated Xinhua Hospital of Dalian University from January 2012 to December 2017 were retrospective analyzed. Variable selection was performed using univariate analysis combined with Least Absolute Shrinkage and Selection Operator (LASSO) regression and 10-fold cross-validation. After constructing the best logistic regression model, multivariate analysis was conducted to determine the independent risk factors for preoperative LNM in CRC, and a nomogram was developed. The model was internally validated using the Bootstrap method and its predictive performance and clinical utility were evaluated through receiver operating characteristic (ROC) curve, calibration curve, and decision curve analysis (DCA). Results Univariate analysis and LASSO regression with cross-validation identified smoking history, neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), lymphocyte-to-monocyte ratio (LMR), fibrinogen-to-albumin ratio (FAR), and fecal occult blood (FOB) as variables with non-zero coefficients. Multivariate analysis using these factors showed that smoking history (OR=2.669, 95%CI: 1.158-6.150, P=0.021), high NLR(OR=1.895, 95%CI: 1.379-2.605, P＜0.001), low LMR (OR=0.907, 95%CI: 0.823-0.999, P=0.048), high FAR (OR=1.145, 95%CI: 1.062-1.235, P＜0.001), and positive FOB (OR=2.289, 95%CI: 1.132-4.630, P=0.021) were independent risk factors for LNM in CRC (P＜0.05). The ROC curve, calibration curve, and DCA curve indicated that the nomogram constructed in this study provided benefits to patients. Conclusion The risk predictive model constructed in this study demonstrated good predictive performance and clinical utility for preoperatively identifying LNM in CRC patients.