Assessing DDI risks in drug discovery and development: approaches, challenges, and trends

Drug-drug interactions (DDIs) critically influence drug efficacy and safety, posing risks, but also offering therapeutic opportunities in some circumstances. Their dual nature necessitates balanced strategies in drug development, especially for high-unmet-need areas like oncology.

This review explores DDI risk assessment methods, challenges in correlating preclinical data with clinical outcomes, and advancements in predictive tools like physiologically based pharmacokinetic (PBPK) modeling. In particular, recent new publications highlight innovation such as artificial intelligence (AI) on DDI risk prediction and endogenous biomarkers for noninvasive monitoring. A comprehensive literature search was conducted in PubMed for relevant publications up to Oct 25, 2025.

Moving beyond a purely defensive stance, we must strategically manage the dual nature of drug-drug interactions across the entire drug lifecycle. This capability is built on a proactive framework that seamlessly integrates multi-faceted data (computational, in vitro, and clinical) to continuously forecast DDI risks and opportunities. The ultimate endpoint of DDI assessment is its clinical impact, quantified through PBPK-DDI-pharmacodynamics (PD) and PBPK-DDI-Toxicity (Tox) models. Rigorous benchmarking of all these predictive methods remains essential to close the translational gap, enhance R&D efficiency, and advance more viable drug candidates.