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

Area of Science:

• Pharmacology
• Drug Development
• Computational Biology

Background:

• Drug-drug interactions (DDIs) significantly impact drug efficacy and safety, presenting both risks and therapeutic opportunities.
• Managing DDIs is critical in drug development, particularly for high-unmet-need areas like oncology.

Purpose of the Study:

• To review DDI risk assessment methods, challenges in preclinical to clinical correlation, and advancements in predictive tools.
• To highlight innovations in DDI prediction, including artificial intelligence (AI) and endogenous biomarkers.

Main Methods:

• Comprehensive literature search in PubMed up to October 25, 2025.
• Exploration of physiologically based pharmacokinetic (PBPK) modeling and its role in DDI assessment.
• Review of AI and biomarker-based approaches for DDI risk prediction.

Main Results:

• Physiologically based pharmacokinetic (PBPK) modeling, combined with pharmacodynamics (PD) and toxicity (Tox) data, is key for assessing clinical impact.
• Artificial intelligence (AI) and endogenous biomarkers offer novel, noninvasive methods for DDI risk prediction.
• Integrating computational, in vitro, and clinical data into a proactive framework enables continuous forecasting of DDI risks and opportunities.

Conclusions:

• A proactive strategy integrating multi-faceted data is essential for managing the dual nature of DDIs throughout the drug lifecycle.
• Rigorous benchmarking of predictive methods is vital to bridge the translational gap, improve R&D efficiency, and advance drug candidates.
• Advanced predictive tools and integrated data frameworks are key to optimizing drug development and mitigating DDI-related risks.