Kimura, G. | bioRxiv Science

bioRxiv2020

Antifungal effects of PC945, a novel inhaled triazole, on Candida albicans-infected immunocompromised mice

Although Candida spp. are frequently detected in fungal cultures of respiratory secretions, their presence is normally assumed to reflect benign colonization. However, there is growing evidence that Candida spp. are involved in the pathogenesis of respiratory diseases such as bronchiectasis and asthma. The aim of this study is to investigate the in vitro and in vivo effects of a novel antifungal triazole, PC945, optimised for topical delivery, against C. albicans. In temporarily neutropenic, immunocompromised mice, C. albicans (529L [ATCC(R)MYA4901] strain), inoculated intranasally, induced acute lung injury and death, associated with higher fungal burden and cytokine induction in the lung. PC945 saline suspension, dosed intranasally once daily, starting one day post candida inoculation, dose-dependently (0.56 ~ 14 g/mouse) improved survival rate and inhibited fungal load in the lung on Day 5 post inoculation. These effects by PC945 were 7 ~ 25-fold more potent than those of voriconazole, despite being of similar in vitro antifungal activity versus this strain. Furthermore, extended prophylaxis with low dose PC945 (0.56 g/mouse) was found to inhibit fungal load more potently than the shorter treatment regimens, suggesting antifungal effects of PC945 accumulated on repeat dosing. In addition, antifungal susceptibility testing on 88 candida isolates (C. albicans, C. parapsilosis, C. tropicalis, C. lucitaniae, C. glabrata, C. guilliermondii) revealed that PC945 has potent effects on Candida species broadly. Thus, PC945 has the potential to be a novel topical therapy for the treatment of C. albicans pulmonary infection in humans.

pharmacology and toxicology↗

bioRxiv2019

Automated acquisition of knowledge beyond pathologists

Deep learning algorithms have been successfully used in medical image classification and cancer detection. In the next stage, the technology of acquiring explainable knowledge from medical images is highly desired. Herein, fully automated acquisition of explainable features from annotation-free histopathological images is achieved via revealing statistical distortions in datasets by introducing the way of pathologists examination into a set of deep neural networks. As validation, we compared the prediction accuracy of prostate cancer recurrence using our algorithm-generated features with that of diagnosis by an expert pathologist using established criteria on 13,188 whole-mount pathology images. Our method found not only the findings established by humans but also features that have not been recognized so far, and showed higher accuracy than human in prognostic prediction. This study provides a new field to the deep learning approach as a novel tool for discovering uncharted knowledge, leading to effective treatments and drug discovery.

bioinformatics↗