The field of Machine Learning (ML) has garnered significant attention, particularly in healthcare for predicting disease severity. Recently, the pharmaceutical sector has also adopted ML techniques in various stages of drug development. Tablets are the most common pharmaceutical formulations, with their efficacy influenced by the physicochemical properties of active ingredients, in-process parameters, and formulation components. In this study, we developed ML-based prediction models for disintegration time, friability, and water absorption ratio of fast disintegration tablets. The model development process included data visualization, pre-processing, splitting, ML model creation, and evaluation. We evaluated the models using root mean square error (RMSE) and R-squared score (R2). After hyperparameter tuning and cross-validation, the voting regressor model demonstrated the best performance for predicting disintegration time (RMSE: 21.99, R2: 0.76), surpassing previously reported models. The random forest regressor achieved the best results for friability prediction (RMSE: 0.142, R2: 0.7), and the K-nearest neighbor (KNN) regressor excelled in predicting the water absorption ratio (RMSE: 10.07, R2: 0.94). Notably, predicting friability and water absorption ratio using ML models is unprecedented in the literature. The developed models were deployed in a web app for easy access by anyone. These ML models can significantly enhance the tablet development phase by minimizing experimental iterations and material usage, thereby reducing costs and saving time.
Increasing resistance to antiviral drugs approved for the treatment of influenza urges the development of novel compounds. Ideally, this should be complemented by a careful consideration of the administration route. 6′siallyllactosamine-functionalized β-cyclodextrin (CD-6′SLN) is a novel entry inhibitor that acts as a mimic of the primary attachment receptor of influenza, sialic acid. In this study, we aimed to develop a dry powder formulation of CD-6′SLN to assess its in vivo antiviral activity after administration via the pulmonary route. By means of spray drying the compound together with trileucine, a dispersion enhancer, we created a powder that retained the antiviral effect of the drug, remained stable under elevated temperature conditions and performed well in a dry powder inhaler. To test the efficacy of the dry powder drug against influenza infection in vivo, infected mice were treated with CD-6′SLN using an aerosol generator that allowed for the controlled administration of powder formulations to the lungs of mice. CD-6′SLN was effective in mitigating the course of the disease compared to the control groups, reflected by lower disease activity scores and by the prevention of virus-induced IL-6 production. Our data show that CD-6′SLN can be formulated as a stable dry powder that is suitable for use in a dry powder inhaler and is effective when administered via the pulmonary route to influenza-infected mice.