(M-125) Explore the Impact of Pharmacological Target-Mediated Low Plasma Exposure in Lead Compound Selection in Drug Discovery – a Modeling Approach

Guohua An, PhD, MD – Associate Professor of Pharmaceutical Sciences and Experimental Therapeutics, University of Iowa; Duxin Sun, phd – Professor of Pharmacy and Professor of Pharmaceutical Sciences, University of Michigan

Objectives: The high failure rate of small-molecule drug development, primarily due to efficacy and safety concerns, has been widely discussed in drug discovery, including challenges in validating new targets, limited predictive abilities of animal models, etc.1 However, a potential reason that has not been raised is the selection of unsuited lead candidates due to some unideal preclinical pharmacokinetic (PK) criteria, such as the preference of selecting compounds with adequate plasma exposure and half-life as lead candidates. These criteria may lead to the triage of compounds with short half-lives or low plasma exposure, possibly due to tissue target binding. Target-mediated low plasma exposure may occur for compounds binding high-capacity targets as the target won’t be saturated even in preclinical stages, potentially excluding promising candidates when comparing analogs. We aimed to build a minimal physiologically-based pharmacokinetics (mPBPK) model to describe drug PK profiles when binding to high-capacity tissue targets to raise awareness of pharmacological target-mediated low plasma exposure and the potential lead compound mis-selection associated with it.

Methods: NONMEM was used for modeling and simulation. A mPBPK model was built, including plasma, low-perfused tissue, and high-perfused tissue compartments, with first-order association rate constant (Kass) and dissociation rate constant (Kdis) integrated into the high-perfused tissue to depict the drug-target binding process. To assess the impact of changes in target binding (Kass and Kdis) on PK and lead compound selection, simulations were run for 7 virtual analogs for single I.V. doses, and model fittings were conducted for mice plasma and various tissue data of four epidermal growth factor receptor (EGFR) inhibitors.

Results: Simulated results indicate that both Kass and Kdis could affect drug distribution in plasma and target-expressing tissues, but Kass plays a more important role. The compound with the highest target binding (Compound 7, Kass of 30) has the lowest plasma exposure among the 7 analogs. However, exposure in target-expressing tissue correlates positively with Kass: Compound 1, lacking target binding, shows the lowest exposure, while Compound 7, with the highest Kass, shows the highest exposure. Also, the Kass was estimated to be 32, 9.17, 3.22, and 16.5 h-1 for afatinib, gefitinib, sorafenib, and dasatinib, showing a strong correlation with their AUC ratios of target-expressing tissue to plasma (R-squared > 0.99). The results confirm the simulated results that the Kass markedly affects drug distribution.

Conclusions: A mPBPK model was built to depict the PK traits of small molecules binding to high-capacity tissue targets. Both simulated and fitted results suggest that drug binding to high-capacity tissue targets could lead to low plasma exposure, which may mislead the lead candidate selection. Considering this may improve the drug development successful rate.

Citations: [1] Sun D, et al. Acta Pharm Sin B. 2022;12(7):3049-62