Construction of prediction model of lymph node metastasis of early cervical cancer based on machine learning algorithm and its application: experience of 204 cases in a single center.

OBJECTIVES: The prediction model of para-aortic lymph node metastasis (LNM) in patients with early cervical cancer was constructed based on the logistic regression (LR) and random forest (RF) algorithms in the machine learning algorithm. The prediction efficiencies of the two models were compared. METHODS: The clinical data of 204 patients with early cervical cancer in the First Affiliated Hospital of Guangxi Medical University were retrospectively collected. The 204 patients were randomly divided into a training set and a verification set according to a ratio of 3:1. The training set was used to build the model. The verification set was used to evaluate model effectiveness. The para-aortic LNM prediction model of early cervical cancer was established by LR and RF. Receiver operating characteristic curve (ROC), sensitivity, and specificity were used to evaluate the prediction performances of the two models. RESULTS: LR analysis showed that tumor diameter > 4 cm, choroidal aneurysm embolism, pelvic lymph node metastasis, and high preoperative squamous cell carcinoma antigen (SCC-Ag) level were risk factors for para-aortic LNM in patients with early cervical cancer (P < 0.05). The area under the ROC curve (AUC) was 0.914. The sensitivity, specificity, and accuracy were 92.6%, 66.7%, 87.0%, respectively. The results of the importance analysis of the characteristic variables of the RF showed that the top 5 variables were preoperative SCC-Ag level, tumor diameter > 4 cm, advanced clinical stage, cancer thrombus, and pelvic lymph node metastasis. The AUC of the RF was 0.883. The sensitivity, specificity, and accuracy were 90.7%, 53.3%, 82.6%, respectively. There was no significant difference in AUC between the LR and RF (P > 0.05). CONCLUSIONS: Both LR and RF models based on machine learning algorithm have great predictive value in predicting early cervical cancer para-aortic lymph node metastasis.

miR-944 acts as a prognostic marker and promotes the tumor progression in endometrial cancer.

microRNA-944 (miR-944) has been reported to be deregulation and play either tumor suppressive or oncogenic function in human malignancies. Previous studies have reported that miR-944 is overexpressed in gynecological malignancies including cervical cancer and breast cancer. While, the clinical significance of miR-944 and its function in human endometrial carcinoma (EC) keep unknown. The levels of miR-944 were analyzed in 68 EC tissues and 20 normal endometrial tissues. Results showed that miR-944 was significantly overexpressed in EC tissues compared to normal endometrial tissues. In addition, increased levels of miR-944 also observed in EC cell lines. Clinicopathological analysis verified that miR-944 expression was associated with international federation of gynecology and obstetrics (FIGO) stages and pathology classification of EC. Moreover, Kaplan-Meier analysis found that miR-944 highly expressing EC patients showed a notable shorter survival. Further experiments revealed that miR-944 knockdown significantly inhibited growth and cell cycle progression while facilitated apoptosis in Ishikawa cells. In turn, miR-944 overexpression prominently promoted proliferation and cell cycle progression, and suppressed apoptosis in KLE cells. In vivo experiments further confirmed that miR-944 silencing notably restrained the tumor growth of EC in nude mice. Mechanically, cell adhesion molecule 2 (CADM2) was recognized as a direct downstream target of miR-944 in EC cells. An inverse correlation between miR-944 and CADM2 expression was observed in EC tissues. Interestingly, CADM2 overexpression showed similar effects to miR-944 knockdown in Ishikawa cells with decreased growth, cell cycle arrest at G1 phase and increased apoptosis. Taken together, this work support the first evidence that miR-944 can be potentially used as a promising biomarker and novel therapeutic target for human EC.

Exploration of miR-1202 and miR-196a in human endometrial cancer based on high throughout gene screening analysis.

Altered microRNA (miRNA) expression has been reported to participate in the pathogenesis of several human diseases, and particularly cancer. The present study examined the involvement of various miRNAs in the pathophysiology of endometrial cancer (EC) and atypical endometrial hyperplasia (AEH). We performed a high-throughput analysis of the miRNAs (miRNA microarray) found in samples of endometrial tissue obtained from 45 patients; among whom, 15 patients were diagnosed with EC, 15 patients were diagnosed with AEH, and the remainder were healthy donors. Next, we selected several miRNAs which exhibited at least a 2-fold difference in expression with a P<0.05 to validate these changes in 3 independent in vitro experiments that used real-time PCR analysis. Finally, miR-1202 and miR-196a were selected as target molecules whose effects on cell apoptosis, cell cycle changes, cell migratory and invasive abilities were investigated using flow cytometric and Transwell assays, respectively, after pre-treatment in vitro. After analyzing 125 miRNAs in a microarray assay, 6 miRNAs (3-high and 3-low expression) were further evaluated via paired comparison in all 3 groups. The validation test revealed a positive correlation between the microarray results and a high level of miR-1202 and a low level of miR-196a in the EC group, when compared with the AEH group. All of the data were normalized with data obtained from normal control donors. We found that either miR-1202 silencing or miR-196a overexpression affected AN3CA and HEC-1-A cells by increasing their apoptosis level and inducing G1 phase arrest while decreasing their migratory and invasive abilities. Inhibitors of miR-1202 and mimics of miR­196a may exert a protective effect, suggesting that miR-1202 and miR­196a may serve as biomarkers for evaluating the effectiveness of EC treatment.

miR-139-5p regulates proliferation, apoptosis, and cell cycle of uterine leiomyoma cells by targeting TPD52.

BACKGROUND: Uterine leiomyoma is one of the most common benign tumors in women. It dramatically decreases the quality of life in the affected women. However, there is a lack of effective treatment paradigms. Micro-RNAs are small noncoding RNA molecules that are extensively expressed in organisms, and they are interrelated with the occurrence and development of the tumor. miR-139-5p was found to be downregulated in various cancers, but its function and mechanism in uterine leiomyoma remain unknown. The aim of this study was to investigate the role of miR-139-5p and its target gene in uterine leiomyoma. METHODS: By using a bioinformatic assay, it was found that TPD52 was a potential target gene of miR-139-5p. Then, expressions of miR-139-5p and TPD52 in uterine leiomyoma and adjacent myometrium tissues were evaluated by quantitative real-time polymerase chain reaction and Western blot. Proliferation, apoptosis, and cell cycle of uterine leiomyoma cells transfected by miR-139-5p mimics or TPD52 siRNA were determined. RESULTS: It was observed that the expression of miR-139-5p in uterine leiomyoma tissues was significantly lower (P<0.001) than that in the adjacent myometrium tissues. Overexpression of miR-139-5p inhibited the growth of uterine leiomyoma cells and induced apoptosis and G1 phase arrest. Dual-luciferase reporter assay and Western blot validated that TPD52 is the target gene of miR-139-5p. Furthermore, downregulation of TPD52 by siRNA in uterine leiomyoma cells inhibited cell proliferation and induced cell apoptosis and G1 phase arrest. CONCLUSION: Data suggested that miR-139-5p inhibited the proliferation of uterine leiomyoma cells and induced cell apoptosis and G1 phase arrest by targeting TPD52.