Drug Repurposing Then, Now, and in the Future
Hermann AM Mucke · 2022
Drug repurposing goes back much further than the late 2000s when the term first came into broad use. Contrary to the impression that has been created by the publicity enjoyed by a few prominent exceptions, pure serendipity never dominated; scientific considerations were always at the forefront and the field developed as our understanding of molecular pharmacology grew. Ultimately drug repurposing exploits two facts: (a) hardly any drug has only a single molecular target; and (b) modulating a target can have different effects depending on where it is expressed and what the physiological circumstances are. However, “connecting the dots” in new creative ways that allow the right experiments to be designed requires not only comprehensive pharmacological knowledge but also the ability to make mental leaps that transcend conventional thinking. It was this “art-in-science” concept that used to dominate “rational” drug repurposing (as opposed to discovery by high-throughput screening). In recent years computational tools such as in silico docking, data mining and algorithmic expert systems have been refined to the point where they can assist with concept generation, introducing systematic elements that had been largely absent from the field. In combination with advanced network medicine and – above all - molecularly defined disease ontologies instead of organ-based ones the opportunities are almost boundless; recent examples illustrate this.[1] One computational segment where we do not yet bring existing technologies fully to bear is the mining of information that is implicitly contained in sources beyond peer reviewed papers (e.g., regulatory or patent documents, which often contain data that are not readily available elsewhere[2,3]) and their full integration with existing knowledge. This is one of the objectives which the REPO4EU Horizon Europe project will address while developing a drug repurposing framework.