Early derisking of drug targets and chemistry is essential to provide drug projects with the best chance of success. Target safety assessments (TSAs) use target biology, gene and protein expression data, genetic information from humans and animals and competitor compound intelligence to understand the potential safety risks associated with modulating a drug target (1). However, there is a vast amount of information, updated on a daily basis that must be considered for each TSA.
We have developed a data science-based approach that allows acquisition of relevant evidence for an optimal TSA. This is built on expert-led conventional and artificial intelligence-based mining of literature and other bioinformatics databases. Potential safety risks are identified according to an evidence framework, adjusted to the degree of target novelty. Expert knowledge is necessary to interpret the evidence and to take account of the nuances of drug safety, the modality and the intended patient population for each TSA within each project.
Alongside understanding the potential risks associated with inhibiting or activating a drug target, it is key to evaluate the different lead candidates emerging from discovery chemistry to understand their potential for toxicity. This is frequently assessed in early ‘Mini Tox’ studies in the rodent and in the maximum tolerated dose/dose range finding studies (MTD/DRF) studies carried out prior to selecting one drug candidate to go forward to GLP toxicology testing. However, there is a constant drive to move away from animal testing. We have developed a deep generative adversarial network (GAN)-based framework capable of deriving new animal results from existing animal studies without additional experiments. Using pre-existing rat liver toxicogenomic (TGx) data from the Open Toxicogenomics Project-Genomics-Assisted Toxicity Evaluation System (Open TG-GATES), we generated Tox-GAN transcriptomic profiles with high similarity (0.997 6 0.002 in intensity and 0.740 6 0.082 in fold change) to the corresponding real gene expression profiles, proving its utility in gaining a molecular understanding of underlying toxicological mechanisms and gene expression-based biomarker development. To the best of our knowledge, the proposed Tox-GAN model is novel in its ability to generate in vivo transcriptomic profiles for different treatment conditions from chemical structures and holds great promise for generating high-quality toxicogenomic profiles without animal experimentation.
Overall, both Tox-GAN and TSAs take full advantage of the most recent developments in data science and can be used within drug projects to identify and mitigate risks, helping with informed decision making and resource management. These approaches should be used in the earliest stages of a drug project to guide decisions such as target selection, discovery chemistry options, in vitro assay choice and end points for investigative in vivo studies.
1. Roberts, RA (2018) Understanding drug targets: there’s no such thing as bad news. Drug Discovery Today, 23, 1925-1928. https://doi.org/10.1016/j.drudis.2018.05.028
2. Xi Chen, Ruth Roberts, Weida Tong, Zhichao Liu, Tox-GAN: An Artificial Intelligence Approach Alternative to Animal Studies—A Case Study With Toxicogenomics, Toxicological Sciences, Volume 186, Issue 2, April 2022, Pages 242–259, https://doi.org/10.1093/toxsci/kfab157 read more... read less