Pancreatic Neuroendocrine Tumours:Grade Has The Advantage Of Predicting Outcome And Selecting Patients For Chemotherapy:A Retrospective Cohort Study In The SEER Database

Context:Pancreatic neuroendocrine tumours(pNET)are a rare and heterogeneous group of tumours with an increasing incidence.Current staging criteria for pNET remain limited and controversial.Meanwhlie,the impact of chemotherapy on overall survival has not been fully defined.Deep study on the prognostic factors of pNET would improve the level of diagnosis and treatment.Further analysis and improved staging based on big data and the role of chemotherapy and the choice of chemotherapy population in pNET have important research valueObjective:The current study aimed to explore epidemiologic trends of pancreatic neuroendocrine tumours(pNET).To determine feasible improvements to staging criteria and investigate the relationship between chemotherapy and survivalDesignand Participants:A retrospective cohort study design was used to analyse annual cancer incidence rates,patient demographics,tumour site and stage,and treatment of pNET.Data were obtained from the National Cancer Institute's SEER registry for all patients diagnosed with pNET between January 1973 and December 2015.Results:Patients diagnosed after 2010 were more likely to present with age greater than 45 years.T0,T1 status,NO status,MO status,and well differentiation.There has been a steady increase in the recognized incidence of pNET,coincident with the migration to earlier T,N,and M stage and higher levels of differentiation.Multivariate analysis with Cox regression of overall pNET was performed and early year of diagnosis,female,age>45,nonfunctional histologic type,poor and undifferentiated tumor grade,size>2 cm,N1 disease,M1 disease,and extensional disease are found to have a negative effect on survival for pNET.Given the significance of tumor grade/differentiation,we then calculated independent prognostic factors of each grade.In general,M status and surgery are the only two independent prognostic factors of each subcategory in grade.Current AJCC staging criteria was applicable to patients with well differentiation,but not other differentiation.This phenomenon can be explained by the above Cox analysis that tumour grade and N status were significant prognostic metrics and had a relatively high risk ratio in patients with G1 status but not G2 or G3/G4 status.Therefore,the shortcoming of the previous staging is that the Grade classification is not considered.The revised system,defined by Grade,T.N,and M status,could robustly discriminate between survival curves.Chemotherapy was associated with significantly improved survival for patients with poor and undifferentiated tumour grading.Conclusions:Grade is superior to 'T',' N',or 'M' status in predicting outcomes and selecting patients for chemotherapy.It is necessary and feasible to combine grade into current staging criteria.We proposed a modified staging classification defined by Grade,as well as T,N,and M status to improve prodiction of overall survival for patients with pNET.