Randomized exposure-controlled trials; impact of randomization and analysis strategies

Area of Science:

• Pharmacometrics
• Clinical Trial Design
• Drug Development

Background:

• Five potential benefits of randomizing clinical trials by concentration versus dose have been proposed in literature.
• These include increased statistical power, reduced sensitivity to pharmacokinetic variability, robust exposure-response relationships, less biased estimates, and better toxicity control.

Purpose of the Study:

• To investigate the validity of these five proposed benefits when clinical trial results are analyzed using a model-based approach.
• To compare dose-controlled (RDCT), concentration-controlled (RCCT), and biomarker-controlled (RBCT) randomization schemes.

Main Methods:

• Quantitative relationships between dose, concentration, biomarker, and clinical endpoint were defined using pharmacometric models.
• Simulations were performed for RDCT, RCCT, and RBCT schemes.
• Analyses were conducted based on the defined pharmacometric models.

Main Results:

• Concentration-controlled (RCCT) and biomarker-controlled (RBCT) trials showed lower statistical power than dose-controlled (RDCT) trials in model-based analysis.
• Model-based analysis demonstrated increased power for RDCT with higher pharmacokinetic variability.
• Statistical power remained robust to correlations between clearance (CL) and effect concentration (EC50) or maximal effect (Emax).
• Bias in exposure-response estimates was negligible (<3%) across all conditions.
• RCCT and RDCT yielded comparable or differing numbers of adverse events for studies with equal power.

Conclusions:

• The proposed advantages of alternative randomization schemes (RCCT, RBCT) may not materialize when employing model-based analysis.
• Model-based analysis suggests that dose-controlled randomization might be preferable under certain conditions.