Nomogram for the prediction of lymph node metastasis and survival outcomes in rectal neuroendocrine tumour patients undergoing resection.

Background: The current study analysed rectal neuroendocrine tumour (RNET) patients undergoing resection to identify predictive factors and construct nomograms for lymph node metastasis, cancer-specific survival (CSS) and overall survival (OS). Methods: RNET patients registered in the Surveillance, Epidemiology, and End Results (SEER) database were included in this study. Multivariable logistic regression analysis was used to investigate the relationships between clinicopathological factors and lymph node metastasis. A multivariate competing risk model was applied to investigate factors independently associated with CSS. Through the Cox regression model, a multivariable analysis of OS was performed. Nomograms were established based on independent predictive factors. Calibration plots, receiver operating characteristic (ROC) curves and Brier scores were used to evaluate the predictive accuracy of the nomograms. Results: In this study, 1,253 RNET patients were included for further analysis. Tumour size ≥12 mm (P<0.001), T3/T4 stage (P<0.001) and M1 stage (P=0.001) were independently associated with lymph node metastasis. The performance of the nomogram was acceptable for predicting lymph node metastasis, with an area under the ROC curve (AUC) of 0.937 [95% confidence interval (CI): 0.874-1.000]. Calibration curves and the Hosmer-Lemeshow test revealed desirable model calibration (P=0.99996). The multivariate competing risk model analysis showed that grade II (P=0.017), tumour size ≥12 mm (P=0.007), AJCC TNM stage II (P=0.002), stage III (P<0.001) and stage IV (P<0.001) were significantly associated with worse CSS. In the competing risk nomogram model, the time-dependent AUC revealed good discriminatory ability of the model (time from 1 to 107 months, AUC >0.900), and the Brier score showed good accuracy of the nomogram, which was greater than that of the AJCC TNM stage. Multivariate Cox analysis showed that age >60 years (P=0.002), median income ≥$65,000 (P=0.013), AJCC TNM stage III (P=0.038) and AJCC TNM stage IV (P<0.001) were independently associated with worse OS. In the nomogram for the prediction of OS, the C-statistic was 0.703 (95% CI: 0.615-0.792), which was significantly better than that of the AJCC TNM stage (0.703 Conclusions: The present study is the first to establish nomograms with great discrimination and accuracy for the prediction of lymph node metastases, CSS and OS in RNET patients, which can be used to guide treatment decision-making and surveillance.