Comparing the applications of machine learning, PBPK, and population pharmacokinetic models in pharmacokinetic drug–drug interaction prediction

Abstract

The gold‐standard approach for modeling pharmacokinetic mediated drug–drug interactions is the use of physiologically‐based pharmacokinetic modeling and population pharmacokinetics. However, these models require extensive amounts of drug‐specific data generated from a wide variety of in vitro and in vivo models, which are later refined with clinical data and system‐specific parameters. Machine learning has the potential to be utilized for the prediction of drug–drug interactions much earlier in the drug discovery cycle, using inputs derived from, among others, chemical structure. This could lead to refined chemical designs in early drug discovery. Machine‐learning models have many advantages, such as the capacity to automate learning (increasing the speed and scalability of predictions), improved generalizability by learning from multicase historical data, and highlighting statistical and potentially clinically significant relationships between input variables. In contrast, the routinely used mechanistic models (physiologically‐based pharmacokinetic models and population pharmacokinetics) are currently considered more interpretable, reliable, and require a smaller sample size of data, although insights differ on a case‐by‐case basis. Therefore, they may be appropriate for later stages of drug–drug interaction assessment when more in vivo and clinical data are available. A combined approach of using mechanistic models to highlight features that can be used for training machine‐learning models may also be exploitable in the future to improve the performance of machine learning. In this review, we provide concepts, strategic considerations, and compare machine learning to mechanistic modeling for drug–drug interaction risk assessment across the stages of drug discovery and development.

Transformative trends: A comprehensive review on role of artificial intelligence in healthcare and pharmaceutical research

“…These techniques use machine learning algorithms to discover critical features and adjust model parameters, reducing the need for large amounts of data and computer power that has historically been associated with PBPK modelling. 51,52…”

Utility of PBPK Modeling in Predicting and Characterizing Clinical Drug Interactions

“…techniques becomes clearer and more integrated into drug discovery engines, it is likely that PBPK models using AI and/or ML generated data will be used to more rapidly inform on possible DDI outcomes (Chen et al, 2022;Gill et al, 2022). The main contrast between mechanistic dynamic DDI approaches (PBPK or population pharmacokinetic (PopPK) models)…”

“…and AI/ML driven approaches is the ability of the AI/ML models to continually redefine the relationship between input parameters and subsequent DDI outcomes, whereas stand-alone PBPK models still require updating and reprogramming as new data emerges (Gill et al, 2022).…”

Multimodal CNN-DDI: using multimodal CNN for drug to drug interaction associated events

“…A multi-channel feature fusion module was then used to fuse these various features. The study 48 compared ML and PBPK models in predicting drug-drug interactions (DDIs). Data-driven in nature, and able to handle huge datasets with complex relations, ML models are thus well suited for predicting DDIs from disparate databases.…”