Moving beyond Binary Predictions of Human Drug-Induced Liver Injury (DILI) toward Contrasting Relative Risk Potential

Area of Science:

• Hepatotoxicity and Drug Development
• Predictive Toxicology
• Pharmacology

Background:

• Drug-induced liver injury (DILI) is a significant safety concern in drug development.
• Existing methods for predicting DILI have limitations in stratifying risk.
• A need exists for robust tools to assess hepatotoxicity potential of drug candidates.

Purpose of the Study:

• To develop and validate the hepatic risk matrix (HRM) for differentiating DILI risk.
• To assess the predictive performance of HRM in stratifying drug candidates.
• To aid decision-making in early clinical development regarding liver safety.

Main Methods:

• Developed a hybrid HRM scoring system integrating physicochemical properties and toxicity mechanisms.
• Utilized physicochemical models (Rule of Two, Partition Model) and toxicity pathways (cytotoxicity, mitochondrial dysfunction, BSEP inhibition).
• Validated HRM against a curated database of 200 drugs and assessed 28 internal/external drug candidates.

Main Results:

• The hybrid HRM successfully identified 70-80% of drugs with high DILI concern.
• HRM effectively stratified drug candidates based on liver safety outcomes (transaminase elevations, clinical approval).
• Incorporating additional toxicity mechanisms showed minimal improvement in DILI prediction.

Conclusions:

• The hybrid hepatic risk matrix is a valuable tool for portfolio assessment of DILI risk in early drug development.
• HRM aids in critical decisions regarding drug candidate progression, dose selection, and due diligence.
• This stratified approach supports the selection of compounds with lower hepatotoxicity risk.