Selection of doses for phase II clinical trials based on pharmacokinetic variability consideration

Pharmacokinetic variability is an important component of the total variability in drug response, but Phase II dose-response trials frequently are designed without considering this important factor. Mixed-effects model simulation was performed to examine overlap of patient area under the concentration-time curve (AUC) values between doses for drugs with differing inter- and intrapatient pharmacokinetic variability. Based on the results of this simulation, a dose increment of at least threefold is needed to ensure that drug exposure does not overlap in at least 50% of the patient population for a drug that exhibits greater than 25% variability. In contrast, an increment factor of 2 is normally sufficient to produce the same degree of resolution when the variability is less than 25%. These results suggest that a more aggressive choice of administration increments could lead to a better separation in systemic drug exposure between doses. This needs to be balanced against the therapeutic window of an individual drug product.