Abstract

The design and development of a drug is a very long process that generally takes many years of research. The laborious and expensive drug development pipeline, still characterized by a low success rate, requires several steps, including target identification, hit generation, hit-to-lead optimization, and preclinical/clinical evaluation. However, assessment of preclinical safety and potential efficacy in clinical trials are the Achilles’ heels of the study, as the most candidates halt their race to market due to pharmacokinetic (PK) issues. The hit-to-lead optimization, which represents a crucial step in the early drug discovery process, aims at improving the drug-likeness by modifying physicochemical properties, improving pharmacokinetic/pharmacodynamic (PD) profiles, and reducing off-target activities. Increasingly, lead optimization may benefit from the support of in silico studies for target identification and validation, as well as for Quantitative Estimation of Drug-Likeness (QED) predictions (Bickerton et al., 2012) and the adsorption, distribution, metabolism, excretion, and toxicity (ADMET) score (Guan et al., 2019) by adopting machine learning algorithms based on different molecular descriptors which allow to discriminate drug-like and non-drug-like compounds with high accuracy of 90%. Artificial Intelligence and -omic sciences can speed up the drug discovery process, but they should be combined with innovation, effectiveness, and efficiency for the construction of new organic molecules, always trying to follow the twelve principles of Green Chemistry (Anastas and Eghbali, 2010). However, daily, academia and large pharmaceutical companies have to deal with two major problems in their laboratory routines: high quantities of hazardous chemical waste in the production and purification processes and long reaction times. The environmentally friendly synthetic approaches, including mechanosynthesis, sonosynthesis, electrosynthesis, and microwave assisted organic synthesis (MAOS), and combinatorial and automated synthetic methods allow to overcome these issues, enabling greener chemistry in a shorter timescale. For example, MAOS in the synthesis of organic compounds, polymers, inorganic materials, and nanomaterials allows to obtain the desired molecules selectively, in higher yields, and in a few minutes compared to conventional reflux heating, by monitoring parameters like temperature, pressure, and ramping of temperature, and choosing not flammable or corrosive and non-volatile media solvents (Raimondi et al., 2019). OPEN ACCESS