Improving the Working Models for Drug-Drug Interactions: Impact on Preclinical and Clinical Drug Development

Area of Science:

• Pharmacology
• Drug Development
• Medicinal Chemistry

Background:

• Drug-drug interactions (DDIs) arise from compounds affecting CYP, UGTs, or transporters.
• Polypharmacy increases the need for early DDI risk assessment in drug development.
• Current FDA-recommended static models for pharmacokinetic DDIs are often overly conservative, leading to false positives.

Purpose of the Study:

• To refine the workflow for assessing CYP-mediated DDI risk for Boehringer Ingelheim (BI) proprietary compounds.
• To evaluate the predictive performance of existing DDI models using real-world clinical data.
• To enhance the accuracy of DDI risk prediction in early drug development.

Main Methods:

• Utilized the Drug-drug Interaction Risk Calculator (PharmaPendium) to assess mechanistic static models.
• Correlated model predictions with human pharmacokinetic data from Phase I clinical trials.
• Investigated the impact of incorporating human renal or preclinical total excretion data into static models.

Main Results:

• The standard FDA formula demonstrated good performance in predicting DDIs for BI compounds.
• Integrating human renal or preclinical total excretion data improved the predictive accuracy of mechanistic static models.
• Candidate drugs identified as victims in DDIs showed enhanced prediction with integrated excretion data.

Conclusions:

• Basic static models (BSMs) are valuable for early drug discovery to exclude DDI risks due to minimal data requirements and low false negative rates.
• Mechanistic static models (MSMs) are recommended for compounds requiring further DDI risk evaluation.
• The study provides a refined approach to DDI risk assessment, balancing prediction accuracy and resource efficiency.