The Value of Serum miR-139-3p Expression Level in Predicting Postoperative Survival of Colon Cancer Patients.

BACKGROUND: To explore the value of serum miR-139-3p expression level in predicting postoperative survival of colon cancer patients. METHODS: We selected 158 cases enrolled in our hospital from January 2017 to December 2019. Using real-time fluorescence quantitative PCR, the expression extents of serum miR-139-3p among patients suffering from colon cancer were detected. The enrollment of patients was performed in the high or low miR-139-3p group on the basis of the cutoff value determined by ROC curve analysis. The risk elements influencing the postoperative survival of colon cancer patients were analyzed by the Kaplan-Meier approach and univariate and multivariate Cox regression models. RESULTS: Compared with control group, significantly lower expression level of serum miR-139-3p was shown in colon cancer group (P < 0.05). Its low expression of miR-139-3p was associated with TNM stage, degree of differentiation, tumor sizes, lymph node metastasis and vascular infiltration in patients with colon cancer (all P < 0.05), which was also significantly associated with short survival time of colon cancer patients (P < 0.05). Multivariate Cox regression model analysis displayed that TNM phase, lymph node metastasis and miR-139-3p <2.17 were independent risk elements affecting postoperative survival (P < 0.05). CONCLUSION: The low expression level of miR-139-3p is related to the short survival time of colon cancer patients, and it is expected to be used as a biological indicator to predict the postoperative survival of colon cancer.

[Expression and clinical significance of microRNA-21-3p and microRNA-551-5p in patients with acute pancreatitis].

OBJECTIVE: To investigate the diagnostic and diagnostic values of plasma microRNA-21-3p (miR-21-3p) and miR-551-5p expression in patients with acute pancreatitis (AP). METHODS: A prospective observational study was conducted. AP patients admitted to the Third People's Hospital of Hainan Province from January 1st 2017 to December 31st 2019 were enrolled. The patients were divided into mild acute pancreatitis (MAP) group, moderate severe acute pancreatitis (MSAP) group and severe acute pancreatitis (SAP) group according to their severity. Fasting venous blood was collected from all subjects the day after admission, and real-time quantitative polymerase chain reaction (PCR) as used to detect the expression levels of plasma miR-21-3p and miR-551-5p. Rehabilitation, discharge or death were end points of study. In addition, 50 healthy people in the same period were selected as the control group. The receiver operating characteristic (ROC) curve was used to analyze the value of the expression levels of plasma miR-21-3p and miR-551-5p for the diagnosis and prognosis of SAP. Pearson correlation was used to analyze the relationship between the expressions of miR-21-3p and miR-551-5p in SAP patients. RESULTS: A total of 164 AP patients were enrolled, including 72 MAP patients, 47 MSAP patients and 45 SAP patients. Among the SAP patients, 27 cases survived and 18 cases died. There were no deaths in MAP group and MSAP group. The levels of plasma miR-21-3p and miR-551-5p in AP group were significantly higher than those in control group [miR-21-3p (2-ΔΔCt): 2.17±0.90 vs. 0.65±0.12, miR-551-5p (2-ΔΔCt): 1.80±0.73 vs. 0.42±0.08, both P < 0.01]. The expression levels of plasma miR-21-3p and miR-551-5p in AP patients increased gradually with the aggravation of the disease (F values were 11.635, 10.204 respectively, both P < 0.01), and the expression levels of plasma miR-21-3p and miR-551-5p in SAP group were significantly higher than those in MSAP group and MAP group [miR-21-3p (2-ΔΔCt): 3.16±1.08 vs. 1.85±0.71, 1.70±0.64; miR-551-5p (2-ΔΔCt): 2.63±0.95 vs. 1.52±0.46, 1.36±0.40; all P < 0.01]. ROC curve analysis showed that the area under the ROC curve (AUC) and 95% confidence interval (95%CI) of the joint diagnosis of SAP with miR-21-3p and miR-551-5p were significantly higher than that of miR-21-3p or miR-551-5p alone [0.898 (0.841-0.960) vs. 0.820 (0.763-0.882), 0.806 (0.748-0.867), Z1 = 4.480, Z2 = 4.916, both P < 0.05], and the sensitivity was 90.7% and the specificity was 85.0%. The expression levels of plasma miR-21-3p and miR-551-5p in the death group of SAP patients were significantly higher than those in the survival group [miR-21-3p (2-ΔΔCt): 3.75±1.17 vs. 2.66±0.87, miR-551-5p (2-ΔΔCt): 3.17±1.04 vs. 2.24±0.83, both P < 0.01]. ROC curve analysis showed that the AUC and 95%CI of the combined prediction of death in SAP patients with miR-21-3p and miR-551-5p were significantly higher than that of miR-21-3p or miR-551-5p alone [0.933 (0.875-0.996) vs. 0.856 (0.794-0.917), 0.816 (0.759-0.874), Z1 = 4.395, Z2 = 5.520, both P < 0.05], and the sensitivity was 95.2% and the specificity was 87.5%. Correlation analysis showed a positive correlation between the expression level of plasma miR-21-3p and miR-551-5p in SAP patients (r = 0.827, P < 0.001). CONCLUSIONS: The increased expression of plasma miR-21-3p and miR-551-5p are positively correlated with the severity of AP patients. The combination of the two items has a better value in the diagnosis and prognosis evaluation of SAP.

A neural joint model for entity and relation extraction from biomedical text.

BACKGROUND: Extracting biomedical entities and their relations from text has important applications on biomedical research. Previous work primarily utilized feature-based pipeline models to process this task. Many efforts need to be made on feature engineering when feature-based models are employed. Moreover, pipeline models may suffer error propagation and are not able to utilize the interactions between subtasks. Therefore, we propose a neural joint model to extract biomedical entities as well as their relations simultaneously, and it can alleviate the problems above. RESULTS: Our model was evaluated on two tasks, i.e., the task of extracting adverse drug events between drug and disease entities, and the task of extracting resident relations between bacteria and location entities. Compared with the state-of-the-art systems in these tasks, our model improved the F1 scores of the first task by 5.1% in entity recognition and 8.0% in relation extraction, and that of the second task by 9.2% in relation extraction. CONCLUSIONS: The proposed model achieves competitive performances with less work on feature engineering. We demonstrate that the model based on neural networks is effective for biomedical entity and relation extraction. In addition, parameter sharing is an alternative method for neural models to jointly process this task. Our work can facilitate the research on biomedical text mining.