| ObjectiveThere is no report on the prediction model of colorectal neuroendocrine tumor in China.This study analyzes the related factors affecting the prognosis of colorectal neuroendocrine neoplasms by extracting patient data from the Cancer Database of the National Cancer Institute and construct a nomogram model that can be used to predict the prognosis of patients.MethodUsing SEER * Stat 8.3.6 software to select and extract the case information of patients which with pathologically diagnosed colorectal neuroendocrine neoplasms and complete follow-up information from 2010 to 2015 in the SEER database,Including race,age at diagnosis,gender,marital status,primary tumor location,tumor differentiation degree,tumor histological type,tumor TNM stage,tumor size,surgery,operation mode,survival time,survival status and other information.Perform a single factor analysis on the 12 relevant factors extracted,and include the meaningful factors in the single factor analysis results and the factors that may affect the survival of patients in the multi-factor COX regression analysis.Using R3.6.1 software to build a predictive model of independent risk factors affecting patient survival from the results of multivariate analysis and evaluate this model.ResultsThe incidence of colorectal neuroendocrine neoplasms is generally increasing year by year,and the age of disease is between 12 and 95 years old,the number of patients aged 50-60 is the highest,accounting for 42.46% of the total.Male patients are slightly more than female patients,male to female ratio is 1: 0.98.Among them,78.40% were neuroendocrine tumor,20.76% were neuroendocrine carcinoma and 0.84% were mixed adenoneuroendocrine carcinoma.The number of rectal neuroendocrine tumors is the largest,accounting for 80.36% of the total number,followed by colon neuroendocrine tumors,accounting for 15.10%,and the majority of rectal neuroendocrine neoplasms are neuroendocrine tumors(84.13%),and neuroendocrine tumors account for only 49.00% of colon neuroendocrine neoplasms.The 1-year,3-year,and 5-year survival rates for the population were 90.30%(95% CI: 89.71% to 90.89%),85.30%(95% CI: 84.52% to 86.08%),and 82.1%(95%CI: 81.12% ~ 83.08%).The larger the tumor,the lower its 5-year survival rate,and the prognosis of colorectal neuroendocrine tumors with diameter less than 1 cm was better,the 5-year survival rate was as high as 99.50%,while the 5-year survival rate of colorectal neuroendocrine tumors with diameter greater than 3 cm was only 33.6%.Multivariate analysis results showed that age,primary tumor location,tumor differentiation degree,tumor histology type,N stage,M stage,tumor size,and whether or not surgery were independent risk factors affecting patient survival.The consistency index of the prediction model was 0.967(95% CI: 0.963 ~ 0.971)and the accuracy of the prediction model for 1 year,3 years,and 5 years was tested by Bootstrap self-sampling 1,000 times.The results show that the model has a good consistency between the predicted survival rate and the actual survival rate of the patients,and this model has higher forecast value compared with the 1-year,3-year,and 5-year forecast value of the 7th edition of AJCC TNM staging,and the difference is statistically significant(P<0.001).ConclusionsThere are many prognostic factors in patients with colorectal neuroendocrine neoplasms.Among them,the histological type of the tumor,the degree of tumor differentiation,the size of tumor,whether it is accompanied by distant metastasis of the tumor,and whether surgery has a greater impact on the prognosis of the patient.The prognosis of patients without surgery or the histological type is cancer or poorly differentiated or with distant metastasis is poor.The risk prediction model based on independent risk factors has high prediction value,can estimate the survival rate of patients,and helps clinicians to provide individualized treatment plans for different patients. |
PDF Full Text Request