Analysis Of Risk Factors For The Development Of Gastric Neuroendocrine Tumor In 1 Year

Section one Epidemiological characteristics of gastric neuroendocrine tumorBackground and Objective Neuroendocrine tumors are heterogeneous tumors origin on the peptidrgic neurons and neuroendocrine cells which can occur in the body many organs and tissues, including the gastrointestinal tract, bile duct, liver disease, and pancreas, bronchial and lung, adrenal medulla, paraganglia, thyroid, parathyroid gland and other the neuroendocrine cells of any parts. According to an analysis of the National Cancer Institute’s Surveillance, Epidemiology and End Results database, the incidence of NENs has been rising about 500% in the past 30 years which may be attributed to the improvement of the diagnostic level. Gastroenteropancreatic neuroendocrine tumors (GEP NETs) are origin from digestive tract diffuse neuroendocrine system. The greatest incidence of carcinoids were the gastrointestinal tract (67.5%). In recent years, the incidence of GEP NETs has been steadily increasing, and nearly half of them are malignant. The prevalence rate of GEP NETs is second only to colorectal cancer in digestive tract malignant tumors. Several guidelines of diagnosis and treatment of gastric NETs have been published due to the heterogeneity in biology and in clinical behavior of the tumor.Rindi et alclassified gastric NETs into 3 subtypes of carcinoids; type 1 associated with chronic atrophic gastritis, which shows good prognosis; and type 2 associated with multiple endocrine neoplasia type 1 and Zollinger-Ellison syndrome, which usually shows good prognosis but with a few exceptions showing aggressive behavior. Type 3 refers to sporadic cases associated with the greatest malignancy potential, presenting the poorest prognosis among the 3 types. The 2010 World Health Organization (WHO) classification defines the entire group of tumors as neuroendocrine neoplasms and divides the tumors into NET G1, NET G2, and poorly differentiated neuroendocrine carcinoma (G3) based on mitotic count and Ki-67 index. The definition of each grade is as follows:(1) G1:mitotic count,< 2/10 high power fields (HPFs) and/or≤2%Ki-67 index; (2) G2:mitotic count 2-20/10 HPFs and/or 3%-20% Ki-67 index; (3) G3:mitotic count> 20/10 HPFs and/or> 20% Ki-67 index. For China available information on these cancers is rather limited. Therefore, it necessary for us to learn some detailed data for gastric NETs in China.Methods The data were collected prospectively in 127 patients with histologically comfirmed sporadic gastric NENs from January 1999 to February 2015 in Southern Hospital, Southern Medical University and The First Affiliated Hospital, Sun Yat-sen University. To collect their clinical information including age, gender, locations, clinical syndromes, endoscopic features, histopathological characteristics, metastasis patterns, Ki-67 grading and the development in one year. These patients had been diagnosed as gastric neuroendocrine tumor or gastric neuroendocrine carcinoma by endoscopic or surgical biopsy. To follow up the development of the patients in one year. The patients with the size of the tumor or the number of tumors increased, the tumor recurrence in one year, included in the development group. The remaining patients were included in the without development group.All statistical analyses were made with SPSS 13.0. descriptive analysis were used to deal with all data. Kruskal-Wallis Test were used to deal with the relationship between Ki-67 grading and development in one year of the tumor, P<0.05 was considered had statistical significance. Spearman correlation test was used to evaluated the correlation between Ki-67 grading and the development in one year of the tumor, P<0.05 was considered han statistical significance. To draw the ROC curve of Ki-67 grading predict the advance of the tumor and calculate the area of the cure. To get the sensitivity and specificity according to the Youden index.Results Since 2006, gastric neuroendocrine tumors were found to increase rapidly in the number of cases per year. Most of the tumors were located at gastric fundus and gastric corpus, which were less located in the gastric antrum. Lots of the patients showed upper abdominal pain. The positive expression rate of Syn,CgA and CD56 in neuroendocrine tumors was higher than other makers. The proportion of NET and NEC in gastric neuroendocrine tumors was nearly the same. Gastric neuroendocrine tumors were mainly single tumor. Nearly 50% of the tumors had been diagnosed with the invasion of the gastric mucosa of the serous layer and more than half of them had been found in lymph nodes or distant metastasis. Ki-67 grading between the development group and the non development group by Kruskal-Wallis chi square test, χ2=24.421, P=0.00。Spearman correlation analysis showed that there was a correlation between Ki-67 grading and the development of gastric neuroendocrine tumor in 1 year (r=0.477, P=0.000). The area of the ROC curve of Ki-67 grading predict the advance of gastric neuroendocrine tumors within 1 years was 0.751. Youden index=sensitivity+specificity-1. The greater the Youden index, the greater the authenticity. So in the case of the largest Youden index, the sensitivity of Ki-67 grading was 80%, and the specificity was 68.20%.Concluion Since 2006, ENETS formulated the classification criteria by the use of Ki 67 index, people knowed more about gastric NET. What’s more, The progress of endoscopic technology help people find gastric neuroendocrine tumors easily which made the number of the tumor cases each year increased rapidly, the understanding of gastric neuroendocrine tumors also gradually deepened. Ki-67 index is based on histopathological indicators, we can depend which to predict tumor development. The sensitivity was 80.00%, specificity was 68.20%.Section Two Analysis of risk factors for the development of gastric neuroendocrine tumor in 1 yearBackground and Objective In China, Because most of the hospital can not measure the level of serum gastrin, we can not distinguish the types of gastric neuroendocrine tumor. What’s more, there are some limitations of gastric neuroendocrine tumor Ki-67 grade and TNM stage of tumor prognosis prediction. The surveillance of Gastric neuroendocrine tumors patients remain controversial. This study aimed to evaluate the risk factorsand their predictive value of the progress in one year in gNET patients. A predictive grading method is proposed, which can be used to distinguish between gastric neuroendocrine tumors with high rate of development in 1 years and patients with low development rate of gastric neuroendocrine tumors.Methods A total of 127 gNET patientswith histologically comfirmed sporadic gastric NENs from January 1999 to February 2015 in Southern Hospital, Southern Medical University and The First Affiliated Hospital, Sun Yat-sen University were enrolled in this study. To collect their clinical information including gender, age, the number of the tumor, tumor’s location, tumor invasion of gastric mucosa, metastasis patterns, Ki-67 grading, the tumor’s immunohistochemistryCgA,Syn,CD56 and NSE, Patients whether or not smoking and with or without diabetes, Patients with or without primary tumor resection, patients with or without other tumors, and the development in one year. These patients had been diagnosed as gastric neuroendocrine tumor or gastric neuroendocrine carcinoma by endoscopic or surgical biopsy. To follow up the development of the patients in one year. The patients with the size of the tumor or the number of tumors increased, the tumor recurrence in one year, included in the development group. The remaining patients were included in the without development group.By multiple factor regression analysis the results obtained suggested that the tumor’s size, Ki-67 index, the patients whether resection in primary lesions by surgical as the main influencing factors of gastric neuroendocrine tumor patients tumor with development in one year. Based on the above factors, the Predictive grading was proposed.All statistical analyses were made with SPSS 13.0. Chi square test or t test and multiple regression analysis were used to find the indictors of prediction development between two groups. Kruskal-Wallis Test were used to deal with the relationship between predictive grading and development in one year of the tumor, P<0.05 was considered had statistical significance. Spearman correlation test was used to evaluated the correlation between predictive grading and the development in one year of the tumor, P<.05 was considered han statistical significance. To draw the ROC curve of Predictive grading predict the advance of the tumor and calculate the area of the cure. To get the sensitivity and specificity according to the Youden index. The DeLong method was used to compare the AUC, and the difference of the ROC curve of predictive grading and the ROC curve of the Ki-67 grading was compared.Results The incidence of the tumor development in one year is 54.90%.The average age of 127 patients with gastric neuroendocrine tumor was 56.31+12.232, of which 65.4% were male and 34.6% were female. Abdominal pain was the main manifestation in 48.8% of the patients. The prevalence of diabetes was 7.9%, and the smoking rate was 15.7% in the selected patients with gastric neuroendocrine tumors. Univariate analysis showed that age, Patients with or without primary tumor resection, the size of the tumor, tumor invasion of gastric mucosa, metastasis patterns, Ki-67 index, CgA of the tumor, tumor’s location were the risk factors for the development of gastric neuroendocrine tumor in 1 years. There was a significant difference between the two groups. By multiple regression analysis, compared with control group, tumor with bigger size, higher Ki-67 Index and without surgery to resect the gastric tumor are risk factors for the development of tumor in one year in gNET patients (P< 0.05). And we Get the regression equation:P(Y)=1/[1+e^-(-0.934+0.047 tumor size+0.715 Ki-67grading-2.597 whether resect the gastric tumor)]. The tumors were divided into three grades according to tumor size, Ki-67 grade and surgery.Grade I:According to the regression equation, the development probability of the tumor in 1 years was<25.00%.Grade II:According to the regression equation, the development probability of the tumor in 1 years was≥25.00%,<50.00%.Grade III:According to the regression equation, the development probability of the tumor in 1 years was≥50.00%,<75.00%.Grade IV:According to the regression equation, the development probability of the tumor in 1 years was≥75.00%.The development rate of Grade I in one is 10%(3/30). Grade II is 55.6%(5/ 9). Grade III is 58.3%(14/24). Grade IV is 89.2%(33/37).There were different in three grades of predictive grading using Kruskal-Wallis test (x2=37.315, P=0.000). There was a correlation between predictive grading and tumor development in one year (r=0.637, P<0.000). Predictive grading was positively correlated with the probability of development in one year of the patients, which indicated that the higher the grade, the greater the possibility of tumor development in 1 year. The area of the ROC curve of predictive grading predict the advance of gastric neuroendocrine tumors within 1 years was 0.849. Youden index= sensitivity+specificity-1. The greater the Youden index, the greater the authenticity. So in the case of the largest Youden index, the sensitivity of predictive grading was 85.55%, and the specificity was 68.20%. The DeLong method was used to compare the AUC. the difference of the ROC curve of predictive grading and the ROC curve of the Ki-67 grading was compared, z=2.7495, p=0.0059. The area, sensitivity and specificity of predictive grading ROC curves were higher than those of Ki-67 grading.Conclusions Gastric neuroendocrine tumor’s size, Ki-67 grading, Patients with or without primary tumor resection, all of these factors can more accurately reflect the risk of development in one year of gastric neuroendocrine tumor patients. Our grading standards can reflect the probability of development in one year objectively and accurately. The higher the classification, the higher probability of development is, the more often to check is should be.