Ryan Vargo (1), Adeniyi Adewale (2), Jaap Mandema (3), Thomas Kerbusch (1), Andrew Tershakovec (1), Greg Winchell (1)
(1) Merck & Co. (2) Alcon Laboratories (3) Quantitative Solutions
Objectives: The two intermediate doses of a fixed dose combination in development had low Cmax exposures that did not meet bioequivalence criteria. This work: (1) characterizes the impact of changes in exposure between the test and reference products on efficacy, (2) informs the impact of dosing regimen and formulation on the efficacy of reference products across the class to evaluate the relevance of Cmax and AUC variation.
Methods: Pharmacokinetic exposures were modeled to efficacy in a two stage process because the majority of the reference literature data are dose response. A dose exposure model was developed from the individual exposure data for the reference product. The dose exposure model was simulated to produce dose distributions with the individual exposure data from the test product as input. The dose response model was built across the class of compounds from study level literature data across 245 trials including over 106,000 patients. During development of the dose response model, the impact of dosing regimen and formulation on efficacy was evaluated. The dose distributions generated from the dose exposure model were used as input in the dose response model simulations to evaluate the impact on efficacy.
Results: The dose exposure model of the reference drug had a nonlinear slope for Cmax (1.31, 95% CI (1.24, 1.37)) and for AUC (1.07, 95% CI (1.01, 1.13)). Twice a day dosing and extended release formulations have blunted peak exposure when compared to the similar total daily dose of immediate release formulation. The dose response model concluded that twice daily dosing and extended release formulations had better efficacy when compared to the similar total daily dose of immediate release formulation. This result is consistent with the literature data that Cmax is not clinically relevant for efficacy suggesting that concentration need to be maintained above a certain threshold for most of the day, which is more closely related to AUC. Using the dose distributions generated from simulation of the dose exposure model with the test product AUC and Cmax exposures, the absolute difference in efficacy change from baseline versus the references ranged from 0.24-0.65% and 0.42-1.12%, respectively.
Conclusion: The results predict that even with the most conservative assumption that Cmax is the driver for efficacy, the efficacy difference between the products is not clinically relevant.
Reference: PAGE 21 () Abstr 2609 [www.page-meeting.org/?abstract=2609]
Poster: Other Drug/Disease Modelling