Serum MiR-1290 Combined With Texture Analysis Based On Apparent Diffusion Coefficient Images Predict The Treatment Response Of Esophageal Squamous Cell Cancer

Part 1 : Study of Predictive Model for Chemo-radiosensitivity of Esophageal Squamous Cell Cancer Based on Serum mi R-1290 combined with Texture Analysis of Diffusion-weighted MR ImagingObjective:The objective of this study was to explore the clinical potential of texture analysis based on apparent diffusion coefficient(ADC) images combined with serum microRNA-1290( miR-1290) for predicting the therapeutic response of esophageal squamous cell carcinoma(ESCC) to chemoradiotherapy.Methods: The dataset comprised pretreatment ADC images and miR-1290 expression levels in serum samples from 43 eligible patients with pathological diagnosis of ESCC. All the patients have received the concurrent chemoradiotherapy and the treatment efficiency was measured using the RECIST. The region of interest(ROI) was delineated and 3D reconstruction of the ADC maps were performed. Quantitative image parameters were extracted using four methods, which included the intensity value histogram, intensity–size–zone matrix matrix(GLCM) and gray level gradient co-occurrence matrix(GLGCM). The difference among these textural parameters between different therapeutic sensitivity groups were compared using Mann-Whitney U Test. Then the significant different parameters were statistically filtered to identify a subset of reproducible and nonredundant parameter. At last, the expression level of serum miR-1290 and(or) texture parameters were used to construct the predictive model by applying the k-nearest neighbors(kNN: neighbors, 5; distance metric, Euclidean) and artificial neural network(ANN: number of hidden layers, 1) algorithms. Mc Nemar’s test was used to test the statistical difference between the performances of the predictive models in predicting the treatment response.Results: 29(PR+CR) out of 43 patients were classified as responders, while 14(SD + PD) were classified as nonresponders. According to the results of Mann-Whitney U Test, 15 texture parameters extracted from 2D images and 18 from 3D had significant difference between the two groups. Supervised learning models based on parameters extracted from the 3D images(ANN: 76.7%,k-NN:79.1%) showed better classification performance than those extracted from the 2D images(ANN: 65.1%,k-NN:67.4%). The accuracy of the predictive model based on the 3D images texture parameters combined with miR-1290 was 90.7% higher than that of miR-1290 expression level only(69.8%). No statistical difference was observed in the performance of the two classifiers.Conclusion: Both of the texture parameters extracted from ADC images and miR-1290 expression level in serum could act as potential biomarkers of tumor response to chemoradiotherapy in ESCC patients. The predictive model constructed by 3D images texture parameters combined with miR-1290 might serve as a new radiological analysis tool for treatment prediction Part 2: miR-1290 Promotes Cancer Progression by Targeting nuclear factor I/X(NFIX) in esophageal squamous cell carcinomaObjective: The objective of this study was to investigate the biological role of miR-1290 in esophageal squamous cell carcinoma(ESCC) progression and its underlying mechanism.Methods: miR-1290 and NFIX protein expression levels in 40 clinical samples were determined by quantitative RT–PCR and western blot. The role of NFIX as a target gene of miR-1290 was identified using a dual-luciferase reporter assay. The effects of ectopic miR-1290 expression on cell proliferation, migration and invasion were determined using mi R-1290 mimic(inhibitor)-transfected cells in a colony formation or transwell assay. Furthermore, the consequences of NFIX targeting by miR-1290 for cell proliferation and cell cycle regulation were examined in vitro by colony-forming assay and flow cytometry.Results: miR-1290 was significantly up-regulated in ESCC tissue samples than that in the normal adjacent tissues, and its expression level was associated with the tumor invasion and clinical TNM stage ESCC patients. NFIX was identified as a target gene of miR-1290 in ESCC cell lines by dual-luciferase assay, and over-expression of miR-1290 significantly downregulated the mRNA and protein levels of NFIX. Functional experiments showed that overexpression of miR-1290 promoted cell proliferation, G1/S phase transition, and cell migration and invasion in vitro. Moreover, the restored expression of NFIX protein in cells transfected with miR-1290 could attenuate the suppressive effects of miR-1290 on cell proliferation, migration and invasion.Conclusion: Our findings provide new insights into the role of miR-1290 in the development of ESCC, and implicate the potential application of miR-1290 in cancer therapy.