Risk and prognostic nomograms for colorectal neuroendocrine neoplasm with liver metastasis: a population-based study

• Published in: International journal of colorectal disease Vol. 36; no. 9; pp. 1915 - 1927
• Main Authors: Ding, Xueliang, Tian, Shaobo, Hu, Jia, Wang, Guobin, Yu, Xiangnan, Fu, Daan, Yuan, Ye, Wang, Lin, Wang, Zheng
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.09.2021; Springer; Springer Nature B.V
• Subjects: Carcinoembryonic antigen; CEA (Oncology); Decision making; Diagnosis; Epidemiology; Gastroenterology; Health aspects; Hepatology; Internal Medicine; Liver; Medical prognosis; Medical research; Medicine; Medicine & Public Health; Medicine, Experimental; Metastases; Metastasis; Neuroendocrine tumors; Nomograms; Original Article; Patients; Population studies; Population-based studies; Proctology; Prognosis; Regression analysis; Risk factors; Surgery; Survival; Prognostic factor; Colorectal neuroendocrine neoplasms; Nomogram; Liver metastasis; Risk factor
• DOI: 10.1007/s00384-021-03920-y
• ISSN: 0179-1958; 1432-1262

Purpose Liver metastasis (LM) significantly shortens the survival time of colorectal neuroendocrine neoplasms (NENs) patients. This research aimed to explore risk and prognostic factors in colorectal NENs patients with LM and develop nomograms for predicting the risk of LM and survival probability quantitatively. Methods A total of 9926 colorectal NENs patients registered in the Surveillance, Epidemiology, and End Results (SEER) database between 2010 and 2017 were included. Risk factors for LM in colorectal NENs patients were identified by multivariate logistic regression analysis. Potential prognostic factors for colorectal NENs patients with LM were identified by multivariable Cox regression analysis. Nomograms were constructed for quantifying the probability of LM occurrence and survival. Results At diagnosis, 8.7% of colorectal NENs patients suffered LM, with 1-, 3-, and 5-year cancer-specific survival (CSS) rates of 44.3%, 26.5%, and 18.0%, respectively. Factors associated with LM occurrence included gender, age at diagnosis, primary tumor location, carcinoembryonic antigen (CEA), histological differentiation, T stage, and N stage. Age at diagnosis, race, histological differentiation, N stage, tumor size, primary tumor location, primary site surgery, and extraliver metastasis were prognostic factors of cancer-specific mortality. The area under the receiver operating characteristics (ROC) curve of the nomogram for predicting LM was 0.888 (95% CI: 0.877–0.898), and the C-index of the nomogram for estimating CSS probability was 0.705 (95% CI: 0.682–0.729). Conclusions This research identified the risk and prognostic factors in colorectal NENs patients with LM. The nomograms constructed by this study can be convenient tools for facilitating clinical decision-making.