Objective:To investigate the risk factors of deep vein thrombosis in patients with malignant tumors and to construct the predication model.Methods:A retrospective case-control study was conducted. The patients with malignant tumors in the Medical Information Mart for Intensive Care (MIMIC)-Ⅳ database from 2008-2019 were enrolled. The basic information, laboratory examination indexes, complications, disease score and survival time of patients were collected. Patients were classified into thrombotic group and non-thrombotic group based on whether deep vein thrombosis occurred. The differences in the baseline data, laboratory examination indexes and survival status of the 2 groups were compared. The variables with statistically significant differences ( P < 0.05) were included in univariate logistic regression model for the development of deep vein thrombosis, and multivariate logistic regression analysis was further performed by backward method to screen out independent factors influencing the development of deep vein thrombosis, and a prediction model for deep vein thrombosis in patients with malignant tumors was built based on the factors. Taking actual development of DVT as the gold standard, receiver operating characteristic (ROC) curve analysis was used to analyze the constructed model in predicting the efficacy of deep vein thrombosis in patients with malignant tumors. Results:A total of 6 699 patients with malignant tumors were finally included, including 3 803 males (56.8%) and 2 896 females (43.2%). The age of admission was (67±13) years old, ranging from 19 to 101 years old; there were 213 cases (3.2%) in the thrombotic group and 6 486 cases (96.8%) in the non-thrombotic group. There were significant differences in body mass index (BMI), proportion of patients with tumor metastasis, simplified acute physiological score Ⅱ (SAPSⅡ), stay length in intensive care unit, 28-day mortality and 1-year mortality between the 2 groups (all P < 0.01). Neutrophil count, platelet count, the ratio of neutrophil to lymphocyte, the ratio of platelet to lymphocyte (PLR), and systemic immune inflammatory index in the thrombotic group were higher than those in non-thrombotic group; while lymphocyte count, prothrombin time (PT), and partial thromboplastin time in the thrombotic group were lower than those in non-thrombotic group, and the differences were statistically significant (all P < 0.01). Multivariate logistic regression analysis showed that increased BMI ( OR = 1.106, 95% CI: 1.070-1.154, P < 0.001) and the decreased PT ( OR = 0.861, 95% CI: 0.781-0.951, P = 0.003), the increased PLR ( OR = 1.001, 95% CI: 1.000-1.002, P < 0.001), the increased SAPSⅡ( OR = 1.043, 95% CI: 1.025-1.062, P < 0.001) were independent risk factors for deep vein thrombosis in patients with malignant tumors. The constructed prediction model: Logit (P) =-5.566+0.101×BMI-0.149×PT+0.001×PLR+0.042×SAPSⅡ. ROC curve analysis showed that the area under the curve of the constructed model for predicting deep vein thrombosis in patients with malignant tumors was 0.801 (95% CI: 0.743-0.859), which was better than that of a single indicator (the range of area under the curve of the four indexes: 0.654-0.728). The sensitivity and specificity of the constructed model at the optimal cut-off value (-2.86) was 69.8% and 81.7%, respectively. Conclusions:The BMI, PT, PLR and SAPSⅡ are independent influencing factors of deep vein thrombosis in patients with malignant tumors. The model constructed by the above four variables shows a good predictive value in the development of deep vein thrombosis of patients with malignant tumors.

Objective:To investigate the risk factors of deep vein thrombosis in patients with malignant tumors and to construct the predication model.Methods:A retrospective case-control study was conducted. The patients with malignant tumors in the Medical Information Mart for Intensive Care (MIMIC)-Ⅳ database from 2008-2019 were enrolled. The basic information, laboratory examination indexes, complications, disease score and survival time of patients were collected. Patients were classified into thrombotic group and non-thrombotic group based on whether deep vein thrombosis occurred. The differences in the baseline data, laboratory examination indexes and survival status of the 2 groups were compared. The variables with statistically significant differences ( P < 0.05) were included in univariate logistic regression model for the development of deep vein thrombosis, and multivariate logistic regression analysis was further performed by backward method to screen out independent factors influencing the development of deep vein thrombosis, and a prediction model for deep vein thrombosis in patients with malignant tumors was built based on the factors. Taking actual development of DVT as the gold standard, receiver operating characteristic (ROC) curve analysis was used to analyze the constructed model in predicting the efficacy of deep vein thrombosis in patients with malignant tumors. Results:A total of 6 699 patients with malignant tumors were finally included, including 3 803 males (56.8%) and 2 896 females (43.2%). The age of admission was (67±13) years old, ranging from 19 to 101 years old; there were 213 cases (3.2%) in the thrombotic group and 6 486 cases (96.8%) in the non-thrombotic group. There were significant differences in body mass index (BMI), proportion of patients with tumor metastasis, simplified acute physiological score Ⅱ (SAPSⅡ), stay length in intensive care unit, 28-day mortality and 1-year mortality between the 2 groups (all P < 0.01). Neutrophil count, platelet count, the ratio of neutrophil to lymphocyte, the ratio of platelet to lymphocyte (PLR), and systemic immune inflammatory index in the thrombotic group were higher than those in non-thrombotic group; while lymphocyte count, prothrombin time (PT), and partial thromboplastin time in the thrombotic group were lower than those in non-thrombotic group, and the differences were statistically significant (all P < 0.01). Multivariate logistic regression analysis showed that increased BMI ( OR = 1.106, 95% CI: 1.070-1.154, P < 0.001) and the decreased PT ( OR = 0.861, 95% CI: 0.781-0.951, P = 0.003), the increased PLR ( OR = 1.001, 95% CI: 1.000-1.002, P < 0.001), the increased SAPSⅡ( OR = 1.043, 95% CI: 1.025-1.062, P < 0.001) were independent risk factors for deep vein thrombosis in patients with malignant tumors. The constructed prediction model: Logit (P) =-5.566+0.101×BMI-0.149×PT+0.001×PLR+0.042×SAPSⅡ. ROC curve analysis showed that the area under the curve of the constructed model for predicting deep vein thrombosis in patients with malignant tumors was 0.801 (95% CI: 0.743-0.859), which was better than that of a single indicator (the range of area under the curve of the four indexes: 0.654-0.728). The sensitivity and specificity of the constructed model at the optimal cut-off value (-2.86) was 69.8% and 81.7%, respectively. Conclusions:The BMI, PT, PLR and SAPSⅡ are independent influencing factors of deep vein thrombosis in patients with malignant tumors. The model constructed by the above four variables shows a good predictive value in the development of deep vein thrombosis of patients with malignant tumors.

Objective:To investigate the clinical characteristics, treatment, and prognostic factors of pancreatic neuroendocrine tumors (pNETs) patients treated with Sunitinib.Methods:The clinical data of pNETs patients from Pfizer Drug Assistance Program of Cancer Foundation of China from April 2013 to November 2017 were retrospectively analyzed. Follow-up and statistical analysis were performed.Results:A total of 235 patients were enrolled, the patients′ overall survival time was between 4 and 252 months, the 3-years and 5-years survival rates were 73.8% and 60.8%, respectively. Univariate analysis showed that factors such as age, Ki-67 index and surgery were associated with the 3-years survival rates of pNETs patients ( P<0.05). Multivariate analysis demonstrated that the age, Ki-67 index and surgery were independent prognostic factors for pNETs patients ( P<0.05). For patients with liver metastases, univariate analysis revealed that surgery was associated with prognosis ( P<0.05). The 5-years survival rate of 124 patients with extending usage of Sunitinib was 53.3%. Conclusion:PNETs are rare tumors with atypical clinical symptoms and the patients often have metastasis at the initiate diagnosis. The age, Ki-67 index and surgery are associated with the prognosis of pNETs patients.

Objective:To investigate the clinical characteristics, treatment, and prognostic factors of pancreatic neuroendocrine tumors (pNETs) patients treated with Sunitinib.Methods:The clinical data of pNETs patients from Pfizer Drug Assistance Program of Cancer Foundation of China from April 2013 to November 2017 were retrospectively analyzed. Follow-up and statistical analysis were performed.Results:A total of 235 patients were enrolled, the patients′ overall survival time was between 4 and 252 months, the 3-years and 5-years survival rates were 73.8% and 60.8%, respectively. Univariate analysis showed that factors such as age, Ki-67 index and surgery were associated with the 3-years survival rates of pNETs patients ( P<0.05). Multivariate analysis demonstrated that the age, Ki-67 index and surgery were independent prognostic factors for pNETs patients ( P<0.05). For patients with liver metastases, univariate analysis revealed that surgery was associated with prognosis ( P<0.05). The 5-years survival rate of 124 patients with extending usage of Sunitinib was 53.3%. Conclusion:PNETs are rare tumors with atypical clinical symptoms and the patients often have metastasis at the initiate diagnosis. The age, Ki-67 index and surgery are associated with the prognosis of pNETs patients.

Objective: To explore the survival difference of patients with colon and rectal neuroendocrine neoplasm (NEN) at different stages. Methods: We identified 8 679 patients with colorectal NEN diagnosed between 1988 and 2014 from the Surveillance, Epidemiology, and End Results (SEER) registry, including 5 437 rectal NEN and 3 242 colon NEN ( 1 681 cecum NEN ). Survival curve was drawn by Kaplan-Meier method. Prognostic factors were analyzed by univariate analysis and multivariate Cox regression model. Results: The ratio of male patients with colon and rectal NEN was similar to female (P=0.095). Rectal NEN patients were younger (P<0.001), more highly differentiated (P<0.001), and with earlier stage (P<0.001). Survival analysis showed that the survival of rectal NEN was superior to that of colon NEN, with 10-year tumor-specific survival rates of 86.8% and 44.8% respectively (P<0.001). Multivariate Cox analysis showed that age, gender, marital status, primary tumor site, grade, stage and surgery were independent prognostic factors of colorectal NEN (all P<0.01). The most important factor was stage (HR=3.531), followed by differentiation grade (HR=1.856). Stratified analysis displayed that the survival of rectal NEN in stage Ⅰ, Ⅱ and Ⅳ were better than those of corresponding stage of colon NEN (all P<0.05), but worse in stage Ⅲ (P=0.012). While the survival of rectal NEN were significantly better than those of colon NEN within all stages after excluding 1681 cases of cecal NEN (all P<0.05). Among the patients with well-differentiated NEN, the survival of rectal NEN in stage Ⅰ, Ⅲ and Ⅳ were better than those of corresponding stage of colon NEN (all P<0.05) while there was no significant difference in stage Ⅱ(P=0.169). For poor-differentiated NEN, only the survival of rectal NEN patients in stage Ⅳ (P=0.001) was significant longer than those of colon NEN, while there was no significant difference in stage Ⅰ, Ⅱ and Ⅲ (stage Ⅰ: P=0.760; stage Ⅱ: P=0.181; stage Ⅲ: P=0.313). Conclusions The survival of NEN patients in colon and rectum is different. Cecum NEN should be considered as a separated tumor for prognostic analysis due to its special clinicopathologic characteristics.