Risks encountered when not adjusting for diurnal variation and food effect in QTcF analysis based on phase I data.

Area of Science:

• Pharmacology
• Clinical Trial Design
• Drug Safety Evaluation

Background:

• Phase I studies increasingly assess drug-induced QT interval prolongation.
• Standard methods may not adequately control for factors like food intake and variable sampling times.
• Accurate concentration-QT analysis is crucial for early drug safety assessment.

Purpose of the Study:

• To evaluate the reliability of standard pre-specified linear models (PLM) versus adjusted PLMs for QT liability assessment.
• To investigate the impact of uncontrolled factors (meal intake, dosing/sampling times) on QT analysis.
• To determine if adjusted models can improve the accuracy of QT liability evaluation in Phase I trials.

Main Methods:

• Simulation analysis of 1000 QTcF-time profiles for a drug with mild QT liability.
• Comparison of an unadjusted PLM with an adjusted PLM accounting for food effect and clock time.
• Evaluation of model performance under various simulated suboptimal study designs.

Main Results:

• The unadjusted PLM showed an inflated negative rate (up to 50%) in suboptimal study designs.
• The adjusted PLM effectively corrected for imbalances in meal intake and dosing/sampling times.
• Adjusted models yielded more conservative and reliable QT liability assessments compared to unadjusted models.

Conclusions:

• Thorough documentation and adjustment for influential factors are essential for reliable QT effect analysis in Phase I trials.
• Adjusted models enhance the accuracy of QT liability assessment, potentially reducing the need for further QT/QTc studies.
• Optimized analysis methods in early phase trials contribute significantly to drug safety and development efficiency.