摘要
Objective To observe the value of combined synthetic MRI and multiplexed sensitivity encoding diffusion weighted imaging(MUSE-DWI)for differentiating benign and malignant endometrial lesions.Methods Data of 112 patients with benign or malignant endometrial lesion confirmed by pathology were retrospectively analyzed.According to pathologic diagnosis,the patients were divided into malignant group(n=72)and benign group(n=40).Synthetic MRI and MUSE-DWI quantitative parameters,including T1,T2,proton density(PD)and apparent diffusion coefficient(ADC)of all lesions were acquired.The clinical data as well as ADC,T1,T2 and PD values of lesions were compared between groups,and those being significantly different between groups were included in univariate and multivariate logistic regression.Then the univariate and combined models were established for differentiating benign and malignant endometrial lesions.The receiver operating characteristic curves were drawn,and areas under the curves(AUC)were calculated to evaluated the diagnostic efficacy of the models,which were compared with DeLong test.Results Patients'age in malignant group were higher than that in benign group(P＜0.05).The length of the maximum diameter was larger,ADC,T2 and PD values were lower in malignant lesions than those in benign ones(all P＜0.05),while no significant difference of T1 value was found between groups(P=0.074).The AUC of ADC univariate model was 0.966,and there was no significant difference in AUC(0.970)between the combined ADC+T2+PD model(adjusted P＞0.05),but both higher than AUC of T2 univariate model(0.618),PD univariate model(0.664)and the combined T2+PD model(0.668)(all adjusted P＜0.05).Conclusion ADC univariate model and combined model with other parameters of combined synthetic MRI and MUSE-DWI could be used to effectively differentiate benign and malignant endometrial lesions.
摘要
Objective To assess the value of magnetic resonance imaging compilation(MAGiC)sequence in predicting lympho-vascular space invasion(LVSI)in early cervical cancer.Methods The data of 48 patients with cervical cancer confirmed by pathology were collected retrospectively,and classified into LVSI-positive group(n=29)and LVSI-negative group(n=19)according to postop-erative pathological results.MAGiC sequence images of patients were obtained before injecting contrast agents,then the region of interest(ROI)was delineated along the largest dimension edge of the lesion,and T1,T2 and proton density(PD)values were automatically generated by the software.Predictors were screened by univariate analysis and receiver operating characteristic(ROC)curves were drawn to assess their diagnostic efficacy for predicting LVSI in cervical cancer.Results Significant differences were found in T1 and PD values between LVSI-positive and LVSI-negative groups(P=0.003,P=0.017).There were no significant differences in T2 values between the two groups(P=0.414).The area under the curve(AUC)for T1 and PD values to predict LVSI status were 0.73 and 0.721,respectively.Conclusion LVSI-positive group of cervical cancer has lower T1 and PD values than LVSI-negative group based on MAGiC sequence.The MAGiC sequence has a certain application value for predicting LVSI status in early cervical cancer.
摘要
<p><b>BACKGROUND</b>Studies have shown that the drug resistance of gastric cancer cells can be modulated by abnormal expression of microRNAs (miRNAs). We investigated the role of miR-503 in the development of cisplatin resistance in human gastric cancer cell lines.</p><p><b>METHODS</b>MiR-503 expression was measured by quantitative real-time PCR. MTT (3-(4,5-dimethylthiazol-2-yl)-2,5- diphenyltetrazolium bromide) and clonogenic assays were used to examine changes in cell viability and the drug resistance phenotype of cancer cells associated with upregulation or downregulation of the miRNA. A dual-luciferase activity assay was used to verify target genes of miR-503. Immunohistochemistry, Western blotting analysis, and a flow cytometric apoptosis assay were used to elucidate the mechanism by which miR-503 modulates drug resistance in cancer cells.</p><p><b>RESULTS</b>MiR-503 was significantly downregulated in gastric cancer tissues and several gastric cancer cell lines. Additionally, downregulation of miR-503 in the cisplatin (DDP)-resistant gastric cancer cell line SGC7901/DDP was concurrent with the upregulation of insulin-like growth factor-1 receptor (IGF1R) and B-cell lymphoma 2 (BCL2) expression compared with the parental SGC7901 cell line. An in vitro drug sensitivity assay showed that overexpression of miR-503 sensitized SGC7901/DDP cells to cisplatin. The luciferase activity of reporters driven by IGF1R and BCL2 3'-untranslated regions in SGC7901/DDP cells suggested that IGF1R and BCL2 were both direct target genes of miR-503. Enforced miR-503 expression in SGC7901/DDP cells reduced expression of the target proteins, inhibited proliferation, and sensitized the cells to DDP-induced apoptosis.</p><p><b>CONCLUSION</b>Our findings suggest that hsa-miR-503 modulates cisplatin resistance of human gastric cancer cells at least in part by targeting IGF1R and BCL2.</p>