Construction and validation of nomogram for predicting lymph node metastasis risk in small nonfunctioning pancreatic neuroendocrine tumor patients based on SEER database
Objective To construct a nomogram for predicting lymph node metastasis risk in small nonfunctioning pancreatic neuroendocrine tumor(NF-pNET)patients based on the the Surveillance,Epidemiology,and End Results(SEER)database and validate it.Method The medical records of 1858 small NF-pNET patients from SEER database from 2004 to 2018 were download.The influencing factors of lymph node metastasis risk in small NF-pNET patients were analyzed by univariate and multivariate Logistic regression,and nomogram was constructed based on this.The accu-racy and discriminative ability of the nomogram prediction model were evaluated by calibration curve and receiver operat-ing characteristic(ROC)curve,and the clinical application value of the nomogram prediction model was evaluated by de-cision curve analysis(DCA).According to the nomogram total point(NTP)optimal segmentation value determined by X-tile software,the training group and all patients were divided into low-risk group and high-risk group,respectively,the prognosis of the two groups were evaluated.Result A total of 1858 small NF-pNET patients medical records were downloaded from the SEER database,which 1294 cases were included in the training group and 564 cases were included in the validation group.Multivariate Logistic regression analysis showed that age,tumor size,primary site,and pathologi-cal grade were influencing factors for lymph node metastasis in small NF-pNET patients(P＜0.05).The calibration curve showed that there was moderate consistency between the observed and predicted lymph node metastasis risks in small NF-pNET patients predicted by the training and validation group nomogram prediction model.The ROC curve showed that the area under the curve(AUC)of the training group nomogram prediction model in predicting lymph node metastasis risk in small NF-pNET patients was 0.69,while the validation group was 0.70.The DCA curve showed that the nomo-gram prediction model was more reliable in predicting lymph node metastasis risk in small NF-pNET patients,with high-er net benefits.The NTP optimal segmentation value determined by X-tile software was 135.5,and the training group and whole population were divided into low-risk group and high-risk group,respectively.The results showed that the total sur-vival in the high-risk group was shorter than those of the low-risk group(P＜0.01).Conclusion The nomogram predic-tion model has a high accuracy in predicting the risk of lymph node metastasis in small NF-pNET patients,which may help physicians provide appropriate prevention and treatment measures for high-risk patients with lymph node metastasis.
万方数据
维普
