Application of Mechanistic Mathematical Modeling to Toxicology: Quantitative Systems Toxicology (QST).

Abstract

Quantitative systems toxicology (QST) is emerging as an independent field of model-informed drug development (MIDD) with a focus on predicting toxicity endpoints. To enable toxicological predictions, QST models require incorporation of mechanistic details specific to safety applications including the ability to accurately model supratherapeutic doses and appropriately represent safety endpoints. Unique to the field of toxicology, mechanistic knowledge is often described through the use of adverse outcome pathways (AOPs), which formally represent existing knowledge about mechanisms of toxicity. The toxicities represented by QST models can arise from exaggerated or adverse pharmacological effects of engaging the drug's intended target (on-target toxicity) or from adverse events due to modulation of additional targets beyond the primary target (off-target toxicity). In cases of on-target toxicity, QST models can be considered as a type of Quantitative Systems Pharmacology (QSP) model that incorporates safety biomarkers and often includes simulations performed outside the therapeutic dose range to explore potential adverse consequences of exaggerated pharmacology in a pre-clinical or clinical setting. QST models assessing off-target toxicities can be considered distinct from QSP models in that they are typically applicable across molecules of a given modality which can (and often do) have different primary therapeutic targets. Off-target QST models commonly focus on the interrogation of general (e.g. pan-compound) toxicity mechanisms, often within a specific organ system. It can be difficult to categorize a model as purely QSP or QST (given that some models can be considered as both a QSP and a QST model), and therefore, we encourage readers to refer to a model based on its context of use and application. Thus, throughout this chapter, we refer to models as QST models when the context of use is to understand safety-related questions. To illustrate QST modeling approaches, examples of QST model applications for on-target and off-target toxicities at different stages of the drug discovery and development pipeline are presented and discussed. Additionally, contexts of use, triggers, key objectives, and potential impacts of QST models including the types of decisions QST applications can inform across drug discovery and development are reviewed. The chapter concludes with an overview of key challenges and future perspectives in the field of QST.