ABSTRACT
Objectives: We aimed to establish an effective machine learning (ML) model for predicting the risk of distant metastasis (DM) in medullary thyroid carcinoma (MTC). Methods: Demographic data of MTC patients were extracted from the Surveillance, Epidemiology, and End Results (SEER) database of the National Institutes of Health between 2004 and 2015 to develop six ML algorithm models. Models were evaluated based on accuracy, precision, recall rate, F1-score, and area under the receiver operating characteristic curve (AUC). The association between clinicopathological characteristics and target variables was interpreted. Analyses were performed using traditional logistic regression (LR). Results: In total, 2049 patients were included and 138 developed DM. Multivariable LR showed that age, sex, tumor size, extrathyroidal extension, and lymph node metastasis were predictive features for DM in MTC. Among the six ML models, the random forest (RF) had the best predictability in assessing the risk of DM in MTC, with an accuracy, precision, recall rate, F1-score, and AUC higher than those of the traditional binary LR model. Conclusion: RF was superior to traditional LR in predicting the risk of DM in MTC and can provide a valuable reference for clinicians in decision-making.
ABSTRACT
OBJECTIVE: The abnormal expression of LARP1 and E-cadherin (E-cad) is related to tumor occurrence and metastasis. Our study analyzed the expression of LARP1 and E-cad in intrahepatic cholangiocarcinoma (ICC) and investigated the prognostic value of the two proteins. METHODS: Immunohistochemistry was performed to detect the expression of LARP1 and E-cad in 50 ICC clinical specimens with adjacent normal tissues and 20 normal bile duct tissues. In situ hybridization was performed to analyze the expression of LARP1 mRNA in all the specimens. RESULTS: LARP1 protein and mRNA expression levels in ICC tumor tissues were significantly higher compared with corresponding adjacent normal tissues and normal epithelial tissues (P<0.01). E-cad protein expression in ICC tumor tissues was remarkably lower than that of the adjacent normal tissues and benign bile duct tissues (P<0.01). Correlation analysis demonstrated that LARP1 and E-cad expression levels were significantly related with the tumor-node-metastasis staging and lymph node metastasis (P<0.01), while no correlation was observed with patient age, gender, and tumor size. Moreover, Spearman rank correlation test revealed that LARP1 expression was negatively related to E-cad (P<0.05). More importantly, the patients with higher LARP1 expression or lower E-cad expression had a shorter overall survival postoperatively than those with LARP1 lower expression or E-cad higher expression. Multivariate analysis demonstrated that LARP1 and E-cad were both considered as important prognostic factors for survival time. CONCLUSION: These findings suggest that the abnormal expression of LARP1 and E-cad showed a close relationship with the occurrence and metastasis of ICC, leading to poor prognosis indirectly.
