Alpha-fetoprotein mediated targeting of polymeric nanoparticles to treat solid tumors.

Background: Serious side effects caused by paclitaxel formulation, containing toxic solubilizer Cremophor® EL, and its nonspecific accumulation greatly limit clinical paclitaxel application. Aim: To design paclitaxel-loaded copolymer of lactic and glycolic acids nanoparticles decorated with alpha-fetoprotein third domain (rAFP3d-NP) to increase paclitaxel safety profile. Methods: rAFP3d-NP was obtained via carbodiimide technique. Results: The particles were characterized with high paclitaxel loading content of 5% and size of 280 nm. rAFP3d-NP revealed biphasic profile with 67% release of paclitaxel during 220 h. Increased area under the curveinf and mean residence time values after rAFP3d-NP administration confirmed prolonged blood circulation compared with paclitaxel. rAFP3d-NP demonstrated significant tumor growth inhibition at 4T1 and SKOV-3 models. Conclusion: rAFP3d-NP is a promising delivery system for paclitaxel and can be applied similarly for delivery of other hydrophobic drugs.

3D Mueller matrix mapping of layered distributions of depolarisation degree for analysis of prostate adenoma and carcinoma diffuse tissues.

Prostate cancer is the second most common cancer globally in men, and in some countries is now the most diagnosed form of cancer. It is necessary to differentiate between benign and malignant prostate conditions to give accurate diagnoses. We aim to demonstrate the use of a 3D Mueller matrix method to allow quick and easy clinical differentiation between prostate adenoma and carcinoma tissues with different grades and Gleason scores. Histological sections of benign and malignant prostate tumours, obtained by radical prostatectomy, were investigated. We map the degree of depolarisation in the different prostate tumour tissues using a Mueller matrix polarimeter set-up, based on the superposition of a reference laser beam with the interference pattern of the sample in the image plane. The depolarisation distributions can be directly related to the morphology of the biological tissues. The dependences of the magnitude of the 1st to 4th order statistical moments of the depolarisation distribution are determined, which characterise the distributions of the depolarisation values. To determine the diagnostic potential of the method three groups of histological sections of prostate tumour biopsies were formed. The first group contained 36 adenoma tissue samples, while the second contained 36 carcinoma tissue samples of a high grade (grade 4: poorly differentiated-4 + 4 Gleason score), and the third group contained 36 carcinoma tissue samples of a low grade (grade 1: moderately differentiated-3 + 3 Gleason score). Using the calculated values of the statistical moments, tumour tissues are categorised as either adenoma or carcinoma. A high level (> 90%) accuracy of differentiation between adenoma and carcinoma samples was achieved for each group. Differentiation between the high-grade and low-grade carcinoma samples was achieved with an accuracy of 87.5%. The results demonstrate that Mueller matrix mapping of the depolarisation distribution of prostate tumour tissues can accurately differentiate between adenoma and carcinoma, and between different grades of carcinoma. This represents a first step towards the implementation of 3D Mueller matrix mapping for clinical analysis and diagnosis of prostate tumours.

Impact of distance-based metric learning on classification and visualization model performance and structure-activity landscapes.

This study concerns large margin nearest neighbors classifier and its multi-metric extension as the efficient approaches for metric learning which aimed to learn an appropriate distance/similarity function for considered case studies. In recent years, many studies in data mining and pattern recognition have demonstrated that a learned metric can significantly improve the performance in classification, clustering and retrieval tasks. The paper describes application of the metric learning approach to in silico assessment of chemical liabilities. Chemical liabilities, such as adverse effects and toxicity, play a significant role in drug discovery process, in silico assessment of chemical liabilities is an important step aimed to reduce costs and animal testing by complementing or replacing in vitro and in vivo experiments. Here, to our knowledge for the first time, a distance-based metric learning procedures have been applied for in silico assessment of chemical liabilities, the impact of metric learning on structure-activity landscapes and predictive performance of developed models has been analyzed, the learned metric was used in support vector machines. The metric learning results have been illustrated using linear and non-linear data visualization techniques in order to indicate how the change of metrics affected nearest neighbors relations and descriptor space.