The risk factors of lymph node metastasis in early colorectal cancer: a predictive nomogram and risk assessment.

Abstract

Purpose: Endoscopic procedures and surgery are common treatments for early colorectal cancer (CRC). However, only approximately 10% of patients who undergo surgery have lymph node metastases (LNM) detected on postoperative pathology, which often leads to overtreatment. This study aims to comprehensively analyze the risk factors for LNM in early CRC patients, establishing a predictive model to aid in treatment decisions.

Methods: This study reviewed the clinicopathologic data of patients with early CRC who underwent surgery from January 2015 to June 2023. Univariate and multivariate logistic regression analyses were employed to identify LNM risk factors. The receiver operating characteristic (ROC) analysis and calibration curves were also constructed to verify the model's discrimination and calibration. A simplified scale was calculated to promote the risk stratification for LNM.

Results: The study analyzed medical records of 375 patients. Of these, 37 (9.9%) cases had LNM. Univariate analysis identified age, nerve invasion, depth of submucosal invasion, histologic grade, LVI, and tumor budding as risk factors. The multivariate analysis confirmed histologic grade (OR, 13.403; 95% CI, 1.415-126.979; P = 0.024), LVI (OR, 6.703; 95% CI, 2.600-17.284; P < 0.001), and tumor budding (OR, 3.090; 95% CI, 1.082-8.820; P = 0.035) as independent predictors. The optimal nomogram, incorporating six risk factors, demonstrated strong predictability with an area under the ROC curve (AUC) of 0.837 (95% CI, 0.762-0.912). A simplified risk assessment scale with a total score of 19 points was developed.

Conclusion: The study developed a nomogram and a simplified risk assessment scale to predict LNM risk, potentially optimizing the management of early CRC patients.