Computational Prediction of Off‐Target Pharmacology for Discontinued Drugs

Abstract

Most small molecule drugs interact with unintended, often unknown targets that lead to a wide range of undesirable adverse effects in both pre‐clinical and clinical development and that result in safety‐related attrition. Undesired off‐target interactions are generally weak and quite often not detected using current drug discovery assays. Thus, improvement in the early identification of off‐target interactions is critical to reduce the drug attrition rates related to toxicities. In order to better identify potential off‐target interactions that could be linked to predictable safety issues, we have evaluated and designed a novel computational process to predict safety‐relevant low and high‐affinity interactions that are currently not covered in experimental polypharmacology screens. The result of these analyses termed Off Target Safety Assessment (OTSA) assesses more than 1500 mechanisms of action (target + actions). This approach exploits a highly curated training set of about 10 million compounds, with known in‐vitro activities, derived from patents, journals and internally installed external databases. To evaluate the tool, we used a panel of computational tools to predict both primary and secondary pharmacology for 1279 discontinued drug molecules. A total of 33,513 possible interactions were predicted for these drugs, of which 19,513 interactions (i.e. 15 interactions/compound) were associated with known safety warnings/issues at various levels. The top common safety associated off‐targets for these discontinued drugs are i) Adenosine receptor's ii) Opioid receptor's iii) PPAR's iv) PDE's v) Adrenoceptor's and vi) Potassium channel's. The computational OTSA's most often identified the right pharmacological targets, but also predicted a significant number of off‐targets that may be relevant to the observed in‐vivo effects.All authors are employees of AbbVie. The design, study conduct, and financial support for this research was provided by AbbVie. AbbVie participated in the interpretation of data, review, and approval of the publication.