RADIOMICS BASED ON TWO-DIMENSIONAL AND THREE-DIMENSIONAL ULTRASOUND FOR EXTRATHYROIDAL EXTENSION FEATURE PREDICTION IN PAPILLARY THYROID CARCINOMA.

Aim: To evaluate the diagnostic performance of radiomics features of two-dimensional (2D) and three-dimensional (3D) ultrasound (US) in predicting extrathyroidal extension (ETE) status in papillary thyroid carcinoma (PTC). Patients and Methods: 2D and 3D thyroid ultrasound images of 72 PTC patients confirmed by pathology were retrospectively analyzed. The patients were assigned to ETE and non-ETE. The regions of interest (ROIs) were obtained manually. From these images, a larger number of radiomic features were automatically extracted. Lastly, the diagnostic abilities of the radiomics models and a radiologist were evaluated using receiver operating characteristic (ROC) analysis. We extracted 1693 texture features firstly. Results: The area under the ROC curve (AUC) of the radiologist was 0.65. For 2D US, the mean AUC of the three classifiers separately were: 0.744 for logistic regression (LR), 0.694 for multilayer perceptron (MLP), 0.733 for support vector machines (SVM). For 3D US they were 0.876 for LR, 0.825 for MLP, 0.867 for SVM. The diagnostic efficiency of the radiomics was better than radiologist. The LR model had favorable discriminate performance with higher area under the curve. Conclusion: Radiomics based on US image had the potential to preoperatively predict ETE. Radiomics based on 3D US images presented more advantages over radiomics based on 2D US images and radiologist.

Overexpression of a partial human androgen receptor in E. coli: characterization of steroid binding, DNA binding, and immunological properties.

The recent cloning of human androgen receptor (AR) cDNAs in this and other laboratories has provided valuable probes for investigating the structure and function of the AR at the molecular level. We now report the overexpression of a region of the human AR containing both the DNA- and hormone-binding domains in E. coli, which provides a means to produce large amounts of AR for analysis and use in functional studies. Under isopropyl-beta-D-thiogalactopyranoside induction, a tripartite protein, consisting of beta-galactosidase, a collagenase recognition site, and AR polypeptide, was produced in E. coli JM109 using pSS20 a as a vector. About 1 mg of the fused AR could be recovered per liter bacterial culture. The induced protein could readily be detected in a sodium dodecyl sulfate-polyacrylamide gel by Coomassie blue staining. Its identity was confirmed by Western blot analysis using antibodies to both beta-galactosidase and the AR. Scatchard analysis of the androgen-binding activity of the hybrid AR revealed high affinity binding to the synthetic androgen, Mibolerone (Kd, approximately 1.2 nM). Competition studies demonstrated the fusion protein's specificity for androgens. The hybrid receptor formed immune complexes with human anti-AR serum that sedimented at about 19S in 10-50% linear sucrose gradients containing 0.4 M KCl. Gel band shift assays revealed that the hybrid receptor protein forms specific complexes with a synthetic steroid response element derived from the mouse mammary tumor virus long terminal repeat region. These results demonstrate that the recombinant AR expressed in E. coli possesses many of the functional properties characteristic of DNA- and steroid-binding domains of the native AR.

In situ hybridization of prostate-specific antigen mRNA in human prostate.

Prostate-specific antigen (PSA) mRNA was detected by in situ hybridization utilizing a 428 base pair [35S]-labelled cDNA probe from the 3' noncoding region of the PSA gene. Thirty six fresh surgical specimens were collected from patients undergoing radical retropubic prostatectomy for carcinoma of the prostate. Quantitative analysis of the levels of PSA mRNA in both the benign and malignant tissues was performed using an IBAS 2000 Image Analysis System. The results of this study demonstrated that there is a significant decrease in the expression of PSA mRNA in the carcinoma tissue when compared to the benign epithelium. The average binding (number of silver grains/1 x 10(4) microns. 2) for 20 specimens of malignant epithelium was 475 +/- 161 and 586 +/- 140 for 16 specimens of benign epithelium (p less than 0.05). Eleven patients had both benign and malignant tissue from the same surgical specimen available for study. From these paired specimens, the PSA mRNA expression was also significantly reduced in the malignant epithelium when compared to the benign epithelium, 445 +/- 162 and 588 +/- 135 respectively (p less than 0.005). The PSA protein was detected using a monoclonal antibody to PSA with an immunohistochemical staining technique. The PSA protein expression paralleled the expression of the PSA mRNA in the majority of the tissue sections. Many of the tumor specimens showed a heterogeneous expression of PSA, whereas all of the benign epithelium had a uniform high level of PSA expression. In conclusion, PSA mRNA and protein are located only within the glandular epithelial tissue, the expression of PSA protein parallels that of the PSA mRNA, and both the PSA protein and PSA mRNA are significantly decreased in the malignant epithelium when compared to benign prostatic epithelium.