Predicting pharmacodynamic effects through early drug discovery with artificial intelligence-physiologically based pharmacokinetic (AI-PBPK) modelling

Abstract

A mechanism-based pharmacokinetic/pharmacodynamic (PK/PD) model links the concentration-time profile of a drug with its therapeutic effects based on the underlying biological or physiological processes. Clinical endpoints play a pivotal role in drug development. Despite the substantial time and effort invested in screening drugs for favourable pharmacokinetic (PK) properties, they may not consistently yield optimal clinical outcomes. Furthermore, in the virtual compound screening phase, researchers cannot observe clinical outcomes in humans directly. These uncertainties prolong the process of drug development. As incorporation of Artificial Intelligence (AI) into the physiologically based pharmacokinetic/pharmacodynamic (PBPK) model can assist in forecasting pharmacodynamic (PD) effects within the human body, we introduce a methodology for utilizing the AI-PBPK platform to predict the PK and PD outcomes of target compounds in the early drug discovery stage. In this integrated platform, machine learning is used to predict the parameters for the model, and the mechanism-based PD model is used to predict the PD outcome through the PK results. This platform enables researchers to align the PK profile of a drug with desired PD effects at the early drug discovery stage. Case studies are presented to assess and compare five potassium-competitive acid blocker (P-CAB) compounds, after calibration and verification using vonoprazan and revaprazan.

Keywords: P-CABs; PBPK modeling; PD modeling; artificial intelligence (AI); early drug discovery; machine learning (ML).

Main steps used to predict PK and PD outcomes of the compounds. (Step 1) Use different AI related simulations to predict the compound’s ADME and physiochemical properties. (Step 2) Predict PK outcomes using the PBPK model. (Step 3) PD models are used to predict how changes in drug concentrations affect gastric acid secretion and gastric pH. E/E0 is the relative activity of H⁺/K⁺ ATPase by drug; k_sec is the secretion rate constants for intra-gastric H⁺ concentration; k_out is the elimination rate constant for intra-gastric H⁺ concentration; H_obs is the observed concentration of H⁺; I (Inhibition) is the current antisecretory effect (or current pH level) of the drug; I_max is the maximum possible effect (or maximum pH level) of the drug can achieve; The term (I_max -I) represents how far the current effect is from its maximum potential.

FIGURE 1

Workflow for predicting the PK and PD outcomes of a compound using AI-PBPK platform.

FIGURE 2

Simulation results of vonoprazan: (A) Plasma concentration versus time and comparisons with observations from studies A (Kentaro, 2018), B (Tack et al., 2023), C (Jenkins et al., 2015), D (Mulford DJ et al., 2022) and E (Jenkins and Patat, 2017), before and after calibration (adjustment); (B) Logarithm of simulated vonoprazan plasma concentration versus time before and after calibration (adjustment); (C) Concentration in the stomach after calibration.

FIGURE 3

Simulated (solid lines with colours) and observed plasma concentration (dashed lines with colures) of revaprazan after calibration.

FIGURE 4

Simulated gastric pH over 24 h, comparing to observations from Studies A (Sakurai Y et al., 2015), B (Laine et al., 2022) and C (Suzuki et al., 2018), after calibration.

FIGURE 5

Simulation of five P-CAB compounds of (A) Plasma concentration versus time; (B) Gastric pH over 24 h after orally administration.