Novel nomograms to predict lymph node metastasis and distant metastasis in resected patients with early-stage non-small cell lung cancer

Background: We aimed to formulate and validate nomograms to unravel the significant risk factors associated with lymph node metastasis (LNM) and distant metastasis in early- stage non-small cell lung cancer (NSCLC). Methods: Totally, 22403 pathologic T1 (pT1) or pT2 NSCLC patients were extracted from the Surveillance, Epidemiology, and End Results (SEER) database and were further randomly assorted into training and testing cohorts. Clinicopathologic risk predictors associated with distant metastasis and LNM were investigated through multivariate logistic regression. The LNM and distant metastasis nomogram models integrating all significant variables were internally validated in the testing cohort. The prediction efficiency of nomograms was assessed via the receiver operating characteristics curve (ROC), calibration plots and decision curve analysis (DCA). Results: The LNM nomogram displayed a favorable calibration and discrimination in both training and testing cohorts, with an area under the curve (AUC) of 0.721 (95% CI: 0.711-0.734) and 0.713 (95% CI: 0.699-0.727), respectively. And the encouraging prediction efficiency was also revealed in the training cohort (AUC =0.788, 95% CI: 0.761-0.816) and the testing cohort (AUC =0.765, 95% CI: 0.721-0.808) for the distant metastasis nomogram. The calibration plot revealed an optimal consistency between model prediction and practical observation. Both nomograms were endowed with the optimal clinical utility and benefits compared with conventional clinicopathological indicators. Conclusions: Our nomogram models are a promising tool with robust predictive power to effectively and intuitively predict the occurrence of LNM and distant metastasis in early-stage NSCLC.