Establishment and validation of a prognostic nomogram for patients with colorectal neuroendocrine carcinoma.

BACKGROUND

Colorectal neuroendocrine carcinoma is a relatively rare tumor, for which a prognosis prediction model is lacking. Based on the data from Surveillance, Epidemiology, and End Results (SEER) database and Fujian Cancer Hospital, the study constructed and validated a prognostic nomogram to assess overall survival of patients with colorectal neuroendocrine carcinoma(CRNEC).

METHODS

We extracted data of patients diagnosed with CRNEC from the SEER database. These patients were randomly divided into a training cohort(N = 1425) and an internal validation cohort(N = 612). Data of patients diagnosed with CRNEC in Fujian Cancer Hospital was collected as an external validation cohort(N = 54). A prognostic nomogram was established. The performance of the nomogram was assessed with ROC curve, C-index and calibration curve. Decision curve analysis(DCA) and ROC curve were used to compare the prediction efficacy of nomogram with the seventh edition of the TNM classification of the American Joint Commission of Cancer.

RESULTS

Nine variables were identified as independent predictors. Nomogram were established by the nine variables. AUC of the nomogram in predicting 1-, 3- and 5-year OS were 0.900, 0.912 and 0.915 in training cohort, 0.900, 0.925 and 0.919 in internal validation cohort, 0.900, 0.903 and 0.928 in external validation cohort. C-index were 0.845, 0.854 and 0.837. Calibration curves overlapped well with reference lines. Compared with the AJCC TNM staging system, the nomogram performed more effectively. Patients classified into low-risk and high-risk groups by the nomogram scores and performed well in stratification.

CONCLUSION

The prognostic nomogram established and validated in our study can accurately and effectively predict the prognosis of patients with CRNEC.