2009 - St. Petersburg - Russia

PAGE 2009: Methodology- Design
Thaddeus Grasela

Modeling and Simulation Approach to Pediatric Drug Development

Ted Grasela, Julie Passarell, Elizabeth Ludwig, Jill Fiedler-Kelly

(1) Cognigen Corporation, Buffalo NY

Objectives: Administration of a fixed daily dose of Drug X in children five to ten years of age resulted in lower exposure relative to adults given the recommended dose. Clinical trial simulations were designed to support selection of an optimal pediatric dosing regimen expected to attain sufficient exposure for drug effectiveness.

Methods: Data were obtained from two pediatric Phase III studies.  A total of nine different scenarios based on three alternative dosage regimens (weight-based or 2 fixed dose regimens based on weight), varying regional distribution of study sites in Europe (EU) and US (100%/0%, 80%/20%, and 50%/50%, respectively), and three levels of baseline response were simulated.  Each of the trials was replicated 100 times.  A previously developed population PK model was used to generate an estimate of AUC0-24 for each simulated patient for each of the 3 dosage regimens.  Patients were then randomized to placebo or drug treatment.  A linear pharmacodynamic model was used to predict response at 12 weeks for each simulated patient.  A statistically significant difference (α = 0.05) in the change in efficacy response from baseline between placebo and drug-treated patients was considered a successful trial.

Results: When comparing the trial success rates using the three simulated dosing regimens, minimal differences are observed which is in accordance with the similar exposure estimates for these regimens.  These results suggest that the most convenient regimen from a commercial perspective can be used without adversely affecting patient outcome.  As expected, increasing the baseline response level results in an increase in the clinical trial success rate regardless of the dosing regimen or study site distribution.  When comparing the various region-of-origin distributions, enrollment based on 50% US and 50% EU patients tends to predict slightly higher trial success rates.  This finding may be due to region-specific differences in patient factors.

Conclusion: Overall, these simulations suggest improvements in design of future pediatric clinical trials would facilitate selection of appropriate dosing regimens in children. Furthermore, the application of this approach may support the objectives of the EU Pediatric Regulation. 




Reference: PAGE 18 (2009) Abstr 1581 [www.page-meeting.org/?abstract=1581]
Poster: Methodology- Design
Top