Development of A Population Pharmacokinetic and Pharmacodynamic Model Following A Phase II, Study of MOD-4023 in Growth Hormone Deficient Adults

Program: Abstracts - Orals, Featured Poster Presentations, and Posters
Session: SUN 624-646-Growth: Clinical Trials & Observational Studies
Clinical
Sunday, June 16, 2013: 1:45 PM-3:45 PM
Expo Halls ABC (Moscone Center)

Poster Board SUN-637
Gili Hart*1, Serge Guzy2, Leanne Amitzi3, Eyal Fima1 and Oren Herskovits4
1Prolor Biotech, Nes Ziona, Israel, 2POP_PHARM; INC, Albany, CA, 3PROLOR BIOTECH, Israel, 4PROLOR BIOTECH, Nes Ziona, Israel
Background: 

PROLOR Biotech technology enabled the production of a long-acting hGH  (MOD-4023), which may obviate the need for the numerous injections now required for the treatment of growth hormone deficiency.  This technology is based on a natural peptide, the C-terminal peptide (CTP) of the beta chain of human chorionic gonadotropin (hCG), which provides hCG with the required longevity. A Phase II study was  conducted to assess the safety, tolerability and pharmacokinetics/pharmacodynamics (PK/PD) of MOD-4023 and IGF-1, respectively, in adults with GHD.  A PK/PD model was developed in order to characterize the correlations between the MOD-4023 injected dose, MOD-4023 blood concentrations and IGF-1 response. This model will be used for predictive purposes in clinical trial simulations, and thus will aid in the design of future clinical studies.

Methods:

Phoenix NLME was used for model establishment, while the QRPEM algorithm was used as optimization method. A sequence of nested PK models was generated to choose the optimal PK model based on Phase II PK-PD data.  The final base model was a two compartment model with an extravascular compartment and Tlag. A full population PK/PD covariate analysis was performed using the forward/backward covariate deletion method, which evaluated sex, age, height, weight and initial r-hGH dose as potential covariates, reveling a statistically significant relationship between BMI and EC50.

Results:

Internal checks and external validation procedures were performed in order to test the nature of the model. The external validation showed the ability of the model to properly predict for each patient the IGF-1 levels based on Bayesian estimates during a 16 weeks extension of Phase II.

In addition, PK/PD results from Phase 1 were simulated, permitting evaluation of their suitability to the data. The model was able to fit the individual MOD-4023 and IGF-1 data.

Finally, the established model was used to predict the population time profile of IGF-1 for naïve GH deficient adult patients administered  MOD-4023 on a weekly basis. Simulation suggested that 2.5 mg of MOD-4023 administered weekly, will be able to maintain the patients within the IGF-1 normal range.

Conclusion:

Based on the existing data, the established model is precise and potentially predictive and therefore was used to support the dose selection for the Phase III study and for further extended modeling of the correlation between administered dose and clinical end points.

Nothing to Disclose: GH, SG, LA, EF, OH

*Please take note of The Endocrine Society's News Embargo Policy at http://www.endo-society.org/endo2013/media.cfm