Evaluation of methods for achieving stable INR in healthy subjects during a multiple-dose warfarin study.

Purpose: No consistent method is available for finding stable warfarin maintenance doses and fast stabilization of international normalized ratio (INR) values among healthy subjects in experimental warfarin interaction studies. Using data from an earlier study that targeted a stable INR of 1.5-2.0 to test an interaction, we retrospectively evaluated potential dosing algorithms using all methods available to us to decrease the time needed for INR stabilization, which could be useful for future interaction studies in healthy subjects.

Methods: Published pharmacogenetic and clinical dosing algorithms used to initiate pharmacotherapy with warfarin were applied, predicted doses and actual doses were compared by regression analysis, and concentration-time profiles of S-warfarin were simulated using SimCYP® software.

Results: No demographic variables were significantly associated with time to reach a stable, low-intensity INR in this population of relatively young, healthy subjects. Predicted and actual doses were positively correlated for the pharmacogenetic algorithm, but not for the clinical algorithm. INR levels and S-warfarin concentrations were associated with CYP2C9 and VKORC1 genotypes.

Conclusions: Induction to a pharmacodynamic steady state for warfarin for future multiple-dose warfarin drug-interaction studies in healthy volunteers may be predicted using a pharmacogenetic-based dosing algorithm. Simulations revealed that the desired subtherapeutic INR level may be achieved by reducing the predicted dose by approximately 15%. Further study is needed to assess the applicability of this approach to decrease attrition rates and the time needed to reach INR stabilization.